Incoming call processing method, device and system, electronic equipment and storage medium

Abstract

The invention provides an incoming call processing method, device and system, electronic equipment and a storage medium. Under the condition that the incoming call is detected to accord with the preset intelligent interaction condition, the incoming call is answered, the artificial intelligence voice interaction with the calling user is carried out through the artificial intelligence voice assistant, and whether the incoming call belongs to the crank call is judged according to the artificial intelligence voice interaction content; a paid answering process is executed on the incoming call judged to belong to the crank call, the paid answering process comprises an incoming call reminding step, and the incoming call reminding step comprises the step of triggering incoming call reminding after payment confirmation information corresponding to a calling number is detected, so that the crank call answering threshold can be improved, part of crank calls are intercepted, and the call answering efficiency is improved. The incoming call reminding quantity and the actual call volume of crank calls are effectively reduced; and whether the incoming call belongs to the crank call can be accurately judged according to the artificial intelligence voice interaction content, so that the processing rationality of each incoming call can be improved, and the user experience can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, device, system, electronic device, and storage medium for processing incoming calls. Background Art

[0002] In modern communications technology, mobile phones have become an indispensable part of people's daily lives. With the widespread use of smartphones, users are facing an increasingly serious problem of harassing phone calls. The large number of harassing calls, including those for sales and fraud, has caused great distress and annoyance to people.

[0003] Although many applications and mobile phone manufacturers provide caller identification and filtering functions, the existing caller identification and filtering functions rely on marking harassing calls. However, this method has a large lag and requires a certain number of users to answer harassing calls before they are marked based on cloud data or manually marked by users. These users are still victims of harassing calls, and many users are unwilling to mark their calls, which leads to a higher number of actual victims. In order to avoid their calls being marked and blocked, the callers of harassing calls also continue to develop new numbers to call. As a result, in actual applications, there are still a large number of unmarked harassing calls. For these harassing calls, users often cannot determine whether they are harassing calls before answering them. They can only understand the other party's intentions after answering the call. This not only wastes users' time and energy, but may also bring unnecessary troubles and risks.

[0004] Therefore, it is necessary to provide a technical solution that can reduce the negative impact of harassing calls on users. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to provide a method, device, system, electronic device, and storage medium for handling incoming calls to solve the problem that users are frequently subjected to the negative impact of harassing calls. To solve the above technical problems, the embodiments of the present application provide the following technical solutions:

[0006] One or more embodiments of the first aspect of the present application provide a method for processing an incoming call, applied to a called terminal, including:

[0007] When it is detected that the incoming call meets the preset intelligent interaction conditions, the call is answered and the caller is interacted with through the artificial intelligence voice assistant;

[0008] Determine whether the incoming call is a nuisance call based on the content of the artificial intelligence voice interaction;

[0009] A paid answering process is executed for the incoming call determined to be a harassing call, and the paid answering process includes an incoming call reminder step, and the incoming call reminder step includes triggering the incoming call reminder after detecting the payment confirmation information corresponding to the calling number.

[0010] One or more embodiments of the second aspect of the present application provide an incoming call processing device, applied to a called terminal, including:

[0011] The artificial intelligence voice interaction module is used to connect the incoming call and conduct artificial intelligence voice interaction with the caller through the artificial intelligence voice assistant when it detects that the incoming call meets the preset intelligent interaction conditions;

[0012] A paid answering judgment module is used to determine whether the incoming call is a nuisance call based on the content of the artificial intelligence voice interaction;

[0013] The paid answering execution module is used to execute a paid answering process for the incoming call determined to be a harassing call. The paid answering process includes an incoming call reminder step, and the incoming call reminder step includes triggering the incoming call reminder after detecting the payment confirmation information corresponding to the calling number.

[0014] One or more embodiments of the third aspect of the present application provide a call processing system, including: a called terminal and a service terminal;

[0015] The called terminal is in communication with the service end;

[0016] The called terminal is used to execute the incoming call processing method provided by the first aspect of the present application;

[0017] The service end is used to provide service support for the called terminal to execute the incoming call processing method.

[0018] One or more embodiments of the fourth aspect of the present application provide an electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in the first aspect of the present application.

[0019] One or more embodiments of the fifth aspect of the present application provide a computer-readable storage medium having computer-readable instructions stored thereon, which can be executed by a processor to implement the method described in the first aspect of the present application.

[0020] The first aspect of the present application provides a call processing method, which is applied to the called terminal. When it is detected that the incoming call meets the preset intelligent interaction conditions, the incoming call is connected and artificial intelligence voice interaction is performed with the calling user through an artificial intelligence voice assistant. The incoming call is judged to be a harassing call based on the content of the artificial intelligence voice interaction. A paid answering process is performed for the incoming call judged to be a harassing call. The paid answering process includes an incoming call reminder step. The incoming call reminder step includes triggering the incoming call reminder after detecting the payment confirmation information corresponding to the calling number, thereby raising the answering threshold of harassing calls through the paid answering process. Since the calling user is willing to pay the fee to trigger the incoming call reminder, and is unwilling to pay the fee without triggering the incoming call reminder, it is possible to intercept some incoming calls, effectively reducing the number of incoming call reminders and actual call volume of harassing calls; and since answering harassing calls requires the calling user to pay a certain fee to the called user, answering harassing calls can be a good choice for the called user. It brings additional economic compensation, which can effectively alleviate the negative emotions of the called user when answering harassing calls and improve the user experience. In addition, the present application scheme has a high degree of flexibility in intercepting harassing calls, and does not arbitrarily intercept according to the number tag information. If it is an important call, even if it is a call marked as a harassing number by a third-party platform or user, as long as the voice interaction content of the calling user meets the expected intention, or is willing to pay the answering fee, the call will still have a chance to be reminded and answered, thereby retaining the chance of a successful call and avoiding the interception of important calls. In addition, the present application scheme uses artificial intelligence voice interaction to determine whether the call is a harassing call. The method can more accurately identify the calling user's calling intention, thereby accurately determining whether the call is a harassing call. Compared with the traditional method of relying solely on number tags to identify harassing calls, it has higher accuracy, can improve the rationality of handling each call, further meet the user's call answering needs, and improve the user experience.

[0021] The call processing device provided in the second aspect of this application, the call processing system provided in the third aspect, the electronic device provided in the fourth aspect, and the computer-readable storage medium provided in the fifth aspect are based on the same inventive concept as the call processing method provided in the first aspect of this application, and have the same or corresponding beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0023] Figure 1The following schematically illustrates a system architecture diagram of an incoming call processing system provided by some embodiments of the present application;

[0024] Figure 2 The following schematically illustrates a flow chart of a method for handling incoming calls provided in some embodiments of the present application;

[0025] Figure 3 A schematic diagram schematically illustrates an incoming call processing device provided in some embodiments of the present application;

[0026] Figure 4 A schematic diagram schematically illustrates an electronic device provided by some embodiments of the present application;

[0027] Figure 5 A schematic diagram of a computer-readable storage medium provided in some embodiments of the present application is schematically shown. DETAILED DESCRIPTION

[0028] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0029] It should be noted that, unless otherwise stated, the technical terms or scientific terms used in this application should be understood by those skilled in the art to which this application belongs. The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit this application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two, but does not exclude the case of including at least one.

[0030] In this application, "at least one (item)" means one or more, "more than one" means two or more, "at least two (items)" means two or three and more than three, and "and / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c".

[0031] In addition, the terms "first" and "second" are used to distinguish different objects, rather than to describe a specific order. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps, units, or elements is not limited to the listed steps, units, or elements, but may optionally include steps, units, or elements not listed, or may optionally include other steps, units, or elements that are inherent to the process, method, product, or apparatus.

[0032] In view of the problem that the existing technology has poor blocking effect of harassing calls, which causes users to frequently suffer from the negative impact of harassing calls, the embodiments of the present application provide a call processing method, device, system, equipment and computer-readable storage medium to solve the problem that current users frequently suffer from the negative impact of harassing calls. Its main technical concept is to construct a paid answering mechanism through technical means to handle harassing calls. The paid answering mechanism sets a call reminder threshold so that harassing calls that are unwilling to pay cannot trigger reminders, thereby reducing the frequency of users being disturbed; and calls that are willing to pay bring economic benefits to the called users and alleviate the negative emotions of answering the calls. The constructed paid answering solution is, for example, When it is detected that an incoming call meets the preset intelligent interaction conditions, the call is answered and artificial intelligence voice interaction is performed with the calling user through the artificial intelligence voice assistant. Whether the incoming call is a harassing call is determined based on the content of the artificial intelligence voice interaction. A paid answering process is performed for the incoming call determined to be a harassing call. The paid answering process includes a call reminder step. The call reminder step includes triggering a call reminder after detecting the payment confirmation information corresponding to the caller number, thereby raising the threshold for answering harassing calls through the paid answering process. Since the call reminder can be triggered only if the calling user is willing to pay the fee, and the call reminder will not be triggered if the calling user is unwilling to pay the fee, it can be achieved The interception of some incoming calls effectively reduces the number of incoming call reminders and actual call volume of harassing calls; and since answering harassing calls requires the calling user to pay a certain fee to the called user, answering harassing calls can bring additional financial compensation to the called user, thereby effectively alleviating the negative emotions of the called user in answering harassing calls and improving the user experience; in addition, the present application scheme has a high degree of flexibility in intercepting harassing calls, and does not arbitrarily intercept based on number tag information. If it is an important call, even if it is a call marked as a harassing number by a third-party platform or user, as long as the voice interaction content of the calling user meets the expected intention, or is willing to pay to answer the call Fees, the caller still has the opportunity to be reminded and answered, thereby preserving the chance of a successful call and avoiding important calls being intercepted; in addition, the present application solution is to determine whether the incoming call is a nuisance call based on the content of the artificial intelligence voice interaction after the incoming call is connected and the artificial intelligence voice assistant conducts artificial intelligence voice interaction with the calling user. The artificial intelligence voice interaction can more accurately identify the calling intention of the calling user, thereby accurately determining whether the call is a nuisance call. Compared with the method of relying solely on number marking, the accuracy is higher, which can improve the rationality of handling each call, further meet the user's call answering needs, and improve the user experience.

[0033] It should be noted that one of the technical concepts embodied in the paid answering embodiment of the present application is to require the calling user (promised or actually completed) to pay a certain resource (such as currency, tokens, etc.) to the called user before triggering the called terminal's call reminder, thereby achieving the purpose of anti-harassment and paid answering. Therefore, the paid answering involved in the embodiments of the present application can also be referred to as paid answering, paid reminder, paid ringing, paid ringing, etc. for the calling party, and can also be referred to as paid answering, paid reminder, paid ringing, paid ringing, etc. for the called party. The above concepts are essentially the same, with only differences in description angles and naming conventions. Therefore, the above concepts are equivalent concepts in the technical solution of the present application, and are essentially the technical concept of triggering the called terminal to issue a call reminder after the calling user (promised or actually completed) pays a certain fee to the called user. Correspondingly, the resources paid by the calling user to the called user to trigger the called terminal's call reminder can be referred to as answering fees, ringing fees, ringing fees, reminder fees, compensation fees, harassment compensation, etc., and the meanings expressed are essentially the same. Based on the above description, as long as it is the same as the inventive concept of this application, no matter what name or concept it is specifically implemented in, it should be within the scope of protection of this application.

[0034] It is worth noting that the answering charges provided in the embodiments of the present application are different from the existing call charges. In the prior art, call charges refer to the fees that users need to pay to telecom operators when using mobile phones, landlines or other communication services. They are different from the answering charges paid to the called users in this application. The payment objects, payment purposes and functions of the two are different. They belong to completely different technical concepts and should be strictly distinguished and should not be confused.

[0035] Reference below Figure 1 The system architecture corresponding to the embodiments of the present application is exemplified.

[0036] like Figure 1As shown, it schematically shows an exemplary framework diagram of a call processing system provided by some embodiments of the present application. The call processing system includes at least a called terminal (such as a mobile phone) and a server, wherein the called terminal and the server are connected via the Internet. During a call, the calling user uses the calling terminal to initiate a call to the called terminal used by the called user. After detecting the call request, the called terminal first detects whether the incoming call meets the preset intelligent interaction conditions, such as whether it is an unfamiliar number. If the intelligent interaction conditions are met, the called terminal connects the incoming call and conducts artificial intelligence voice interaction with the calling user through an artificial intelligence voice assistant. Then, based on the content of the artificial intelligence voice interaction, it is determined whether the incoming call is a harassing call. For the incoming call determined to be a harassing call, a paid answering process is executed. The paid answering process includes a paid answering interaction step and a call reminder step. The paid answering interaction step includes playing a paid answering prompt message to the calling user to prompt the calling user to pay a certain answering fee before reminding the called user to answer the call. The call reminder step includes triggering a call reminder after detecting payment confirmation information corresponding to the calling number. The above-mentioned server can provide service support for the called terminal or the application installed on the called terminal, such as number tag query, payment information query, etc. In addition, the server can also be connected to the calling terminal through the Internet and provide service support for the calling terminal or the application installed on the calling terminal, such as convenient payment, contactless payment, automatic payment, etc.

[0037] It should be noted that in the above exemplary description, the call can be implemented through a telecommunications network, for example, through a traditional telephone network (PSTN, public switched telephone network), a next-generation network (NGN) or a mobile phone network, etc., which is currently a more commonly used telephone communication method, but in actual applications, the calls and incoming calls involved in the embodiments of the present application can also be implemented through an Internet link, such as a voice chat call service implemented by a voice service based on the Internet Protocol (VoIP, Voice over Internet Protocol) (the corresponding call request is also called a voice chat invitation, a video chat invitation), etc., which is also applicable to the technical solution provided in the embodiments of the present application and is within the scope of protection of this application.

[0038] In the following description of the embodiments, the above system architecture can be referred to for understanding, but it should be noted that the above system architecture is only an example given for ease of understanding. The actual application scenarios are ever-changing and cannot be exemplified one by one. Those skilled in the art can make reasonable changes based on the examples of this application. Without departing from the main purpose of the invention of this application, they should all be within the scope of protection of this application.

[0039] Both the calling terminal and the called terminal can be any electronic device with voice communication function, which can be a mobile terminal or a fixed terminal, such as a mobile phone, a landline (landline), a smart watch, a smart bracelet, a children's phone watch, a tablet computer, a car terminal (smart car machine), smart glasses, etc. In addition, the calling terminal can also be a call center equipment, an outbound automatic dialer, a dedicated outbound terminal, an artificial intelligence outbound equipment, etc. In addition, with the development of technology, other types of new equipment with voice communication capabilities should also be regarded as the application scope of the present invention.

[0040] In this embodiment, the server side can include various forms of computing resources, including but not limited to network hosts, network servers, network server clusters, or cloud computing-based computer collections. Depending on specific needs, the server side can exist in the form of a single entity or multiple entities. It can be further subdivided into multiple more specific functional entities or functional modules based on business or functions to optimize business processing and improve system efficiency. For example, it can include multiple servers, virtual servers, or server clusters for processing different businesses. For example, to better manage user account information, a dedicated management server can be set up; to efficiently store and retrieve user data, a dedicated database server can be configured; to accurately manage user call credit information, a credit management server can be deployed; to safely and conveniently implement payment functions, a payment service server can be established; to ensure the security and reliability of user authentication, an authentication service server can be built, etc. All of the above servers belong to the server side. Through this layered architecture and the functional division of the server side, it is possible to more flexibly respond to different business needs and technical challenges. In particular, each specific server side focuses on a specific task, thereby improving the responsiveness and stability of the entire system. For example, the management server is responsible for handling all operations related to user accounts, such as registration, login, personal information update, etc., to ensure the integrity and consistency of user data; the database server focuses on efficient data storage and fast query, supporting the management and analysis of large-scale data sets; the credit management server is used to evaluate and maintain the user's call credit record, providing reliable support for paid answering services; the payment service server integrates multiple payment gateways and security protocols to ensure the security and transparency of the transaction process; the authentication service server strictly verifies the user's identity to ensure the security and privacy protection of the system. The various service terminals can work together to provide a smooth and secure communication experience for the calling and called users. It should be noted that the embodiments of the present application do not limit the specific architecture of the server. Since the implementation schemes of the server are ever-changing, it is impossible to give examples one by one. Those skilled in the art can flexibly set the server according to actual needs. As long as it is based on the technical concept provided by the embodiments of the present application, executes the method steps provided by the embodiments of the present application, and is used to achieve the invention purpose of the embodiments of the present application, it should be within the scope of protection of this application.

[0041] The calling terminal and called terminal, calling user and called user, calling number and called number, etc. are all divided according to the roles played in the call. During a call, the party that initiates the call request is called the calling terminal, calling user and calling number, while the party that receives the call request is called the called terminal, called user and called number. This division method applies not only to communications between mobile phones, but also to communications between fixed phones (landlines) and mobile phones, as well as any other form of voice or data communications.

[0042] In some implementations of this embodiment, the executor of the call processing method may be the called terminal, or an application installed on the called terminal, such as a call application, an auxiliary application (such as an artificial intelligence voice assistant), or other applications with a call function. The application may be a system application or a third-party application, which aims to reduce harassing calls and reduce the negative impact of harassing calls on called users by providing users with a paid answering function. In the description of the embodiments of this application, for ease of understanding, the example of achieving a paid answering function by installing a first call application on the called terminal is mainly used for explanation. The first call application is any application with a call function.

[0043] like Figure 2 As shown, a method for processing an incoming call is applied to a called terminal. The method for processing an incoming call may include the following steps S101-S103:

[0044] Step S101: When it is detected that the incoming call meets the preset intelligent interaction conditions, the incoming call is answered and artificial intelligence voice interaction is performed with the calling user through the artificial intelligence voice assistant.

[0045] This step aims to accurately identify the nature of the call through intelligent judgment and interaction mechanisms without disturbing the called user, screen out potential harassing calls, and provide a decision-making basis for the subsequent paid answering process. The above-mentioned preset intelligent interaction conditions can be a set of judgment rules predetermined according to user settings or system default rules for triggering the intervention of the artificial intelligence voice assistant for artificial intelligence voice interaction. The judgment rules in the set can be determined according to the characteristics of the call. The main purpose is to preliminarily screen out potential harassing calls, such as unfamiliar calls, unfamiliar calls without number tags, unfamiliar calls with designated number tags (such as marketing calls), etc. This will be explained in detail later and will not be repeated here.

[0046] The artificial intelligence voice assistant is an intelligent program deployed on the called terminal, or an intelligent module integrated into an application (such as a call application). It is implemented based on artificial intelligence (AI) technology and has capabilities such as automatic speech recognition (ASR), natural language processing (NLP) and text to speech (TTS). It can perform real-time voice interaction with the calling user after the call is connected in the background. Among them, speech recognition is to convert the voice signal into text through the acoustic model and the language model, natural language processing includes semantic understanding, intent recognition and context management, and speech synthesis is to convert text into natural speech output. The embodiment of the present application can use existing artificial intelligence technology to construct an artificial intelligence voice assistant that can realize the functions of the present application, or it can use the existing artificial intelligence voice assistant to make appropriate adjustments and changes to apply to the technical solution of the present application, all of which should be within the scope of protection of the present application. Artificial intelligence voice interaction refers to the conversation between the artificial intelligence voice assistant and the calling user. The artificial intelligence voice assistant can communicate with the calling user based on a preset interactive response template or real-time semantic analysis to understand the calling user's call purpose, identity and other key information to further determine whether the call is a nuisance call.

[0047] At the technical implementation level, for example, when the called terminal detects that the incoming call meets the conditions for intelligent interaction, it takes over the incoming call path through the underlying communication protocol (at this time, the user does not need to manually answer the call), establishes a voice channel without initiating a regular call reminder to the user, and activates a lightweight artificial intelligence voice assistant deployed locally on the called terminal to conduct artificial intelligence voice interaction with the calling user. Among them, artificial intelligence voice interaction can be achieved through the following exemplary process: the intelligent voice assistant background connects the incoming call and takes over the call, obtains the voice signal sent by the calling terminal, and then uses the voice recognition module to convert the voice into text. Then, the semantics and intent of the text are analyzed using natural language processing technology. Then, according to the preset interaction template or by calling the corresponding service interface or database, a reply text is generated. Then, the reply text is converted into a voice signal using speech synthesis technology. The synthesized voice signal is then injected into the call link through the audio channel and sent to the calling terminal. Then, the voice signal sent by the calling terminal is received - converted to text - semantic analysis - generated reply text - synthesized reply voice response, and the cycle is repeated to achieve voice interaction with the calling user. It should be noted that the artificial intelligence voice assistant can be deployed locally and complete the artificial intelligence voice interaction process based on local resources to improve interaction efficiency and data security. In addition, in some cases, the artificial intelligence voice interaction process can also be implemented through cloud-based collaborative computing, which can improve the accuracy of the interaction. The embodiments of this application do not limit its specific implementation method.

[0048] Step S102: Determine whether the incoming call is a nuisance call based on the artificial intelligence voice interaction content.

[0049] The called terminal can use the artificial intelligence voice assistant to conduct multiple rounds of conversations with the calling user, for example, asking the caller's identity, the purpose of the call, guiding the calling user to explain specific matters, etc. Correspondingly, the artificial intelligence voice interaction content will include the caller's identity, the purpose of the call, the specific matters of the call, etc. Based on these interaction contents, it can be determined whether the call is a harassing call.

[0050] Step S103: executing a paid answering process for the incoming call determined to be a nuisance call, the paid answering process including an incoming call reminder step, the incoming call reminder step including triggering an incoming call reminder after detecting payment confirmation information corresponding to the calling number.

[0051] The paid answering process refers to a series of steps performed by the called terminal after determining that the incoming call is a harassing call. One of its purposes is to limit the economic cost of harassing calls for calls through the paid answering mechanism, thereby raising the threshold for answering harassing calls and filtering out low-value or malicious calls. The paid answering process at least includes triggering a call reminder after detecting the payment confirmation information corresponding to the calling number. Among them, the payment confirmation information can be confirmation information for paying the answering fee to the first account. The first account is the account corresponding to the called user who uses the called terminal, that is, the called user account. The first account can be bound to the called number to ensure that the call reminder is triggered only when the called user can receive appropriate financial compensation. In addition, the payment confirmation information can be generated by the called terminal based on the payment situation or the promised payment situation, or it can be generated by the server based on the payment situation or the promised payment situation and then sent to the called terminal, both of which can achieve the purpose of the embodiments of the present application. This application is based on a payment confirmation mechanism, which can effectively raise the threshold for answering harassing calls, reduce the frequency of called users answering meaningless and low-value calls, and ensure that only calling users who are willing to bear the answering fees can reach the called users, thereby accurately distinguishing high-value calls from inefficient harassing calls, and avoiding the misjudgment problem caused by traditional interception technology relying on static tags.

[0052] In some embodiments, in order to protect the rights and interests of both the caller and the called party, the caller may first pay the answering fee to a payment platform (which may be one of the service platforms provided by the server), or the payment platform may deduct the answering fee from the payment account corresponding to the calling number, and then transfer the full or partial amount of the answering fee (the server may deduct part of the service fee) to the called user's account after the called user finishes answering the call. In this case, although the answering fee is paid to the called user's account after two stages, it still complies with the technical concept of the embodiment of the present application (the calling user pays the fee for the successful call connection, and the called user receives financial compensation for answering harassing calls). In some embodiments, from the perspective of the calling user, the purpose of its payment is still to pay the called user. From the perspective of the called user, the answering fee it receives still comes from the calling user. Therefore, even if the payment record records the payment made by the calling user to the payment platform, the called user receives the payment from the payment platform. As long as it essentially complies with the inventive concept of the present application, it is also within the scope of protection of the present application. In addition, in some embodiments, the service end or the called terminal can create a paid answering order for each incoming call, a call determined to be a harassing call, or a call for which the caller agrees to pay the answering fee. The paid answering order has a unique identification code such as an order number. The calling user pays according to the paid answering order. Whether the answering fee is paid to the payment platform or to the called user's account, and the payment platform pays the answering fee to the called terminal, it can be traced and integrated according to the paid answering order to form a complete answering fee flow information to prove that the answering fee is paid by the calling user to the called user. In the above description, when there is an intermediate guarantee by the payment platform, the answering fee needs to trigger the call reminder of the called terminal after it is paid to the payment platform. Therefore, in this case, the payment confirmation information can also be the payment confirmation information of the payment of the answering fee to the service end (such as the payment platform), or the payment confirmation information of the payment of the answering fee to the payment platform for the purpose of paying the called user's account. The payment confirmation information can be associated with the paid answering order for this call to ensure that it ultimately pays the answering fee to the called user.

[0053] Incoming call reminder refers to the process in which the called terminal notifies the called user that an incoming call is waiting to be answered through a preset method. It usually includes various forms, such as ringtones, vibrations, screen prompts, etc. The specific method can be determined according to the incoming call reminder method set by the user in the terminal, and this application does not limit it.

[0054] The call processing method provided in the embodiment of the present application connects the call and conducts artificial intelligence voice interaction with the calling user through an artificial intelligence voice assistant when detecting that the incoming call meets preset intelligent interaction conditions. The call is then judged to be a harassing call based on the content of the artificial intelligence voice interaction. A paid answering process is performed for the call determined to be a harassing call. The paid answering process includes a call reminder step. The call reminder step includes triggering a call reminder after detecting payment confirmation information corresponding to the calling number. In this way, at the initial stage of the call, the nature of the incoming call can be preliminarily identified through artificial intelligence voice interaction without the intervention of the called user. Subsequently, if it is determined to be a harassing call, a paid answering process is initiated. In the paid answering process, whether the called terminal issues a call reminder depends on whether the calling user is willing to pay the answering fee. The call reminder is triggered only when it is detected that the calling user agrees or has completed payment of the answering fee. Otherwise, the call reminder is not issued. Therefore, it is possible to intercept some incoming calls, effectively reduce the number of call reminders and actual call volume of harassing calls, and reduce the frequency of user interruptions. And since the harassing calls that can trigger call reminders are willing to pay the answering fee, the called user can obtain additional financial compensation for answering the harassing calls, which can effectively alleviate the negative emotions of the called user when answering harassing calls and improve the user experience. In addition, the present application solution has a high degree of flexibility in intercepting harassing calls, and does not arbitrarily intercept based on number tag information. If it is an important call, even if it is a call marked as a harassing number by a third-party platform or user, as long as the voice interaction content of the calling user meets the expected intention, or is willing to pay the answering fee, the call still has a chance to be reminded and answered, thereby preserving the chance of a successful call and avoiding the interception of important calls. In addition, the present application solution uses artificial intelligence voice interaction to determine whether the call is a harassing call. It can more accurately identify the calling user's calling intention, thereby accurately determining whether the call is a harassing call. Compared with the traditional method of relying solely on number tags to identify harassing calls, it has higher accuracy, can improve the rationality of handling each call, further meet the user's call answering needs, and improve the user experience.

[0055] Taking into account that in actual applications, users often only need to pay to answer harassing calls, and often do not need to pay to answer calls from known contacts such as acquaintances, nor is there any need for artificial intelligence voice interaction. Therefore, this application sets intelligent interaction conditions to selectively perform intelligent interaction on incoming calls. Before step S101, it is detected whether the incoming call meets the preset intelligent interaction conditions, and step S101 is executed if the intelligent interaction conditions are met. The intelligent interaction conditions can be determined based on the characteristics of the incoming call (such as the characteristics of the caller number, the characteristics of the incoming call time, etc.). In some examples, the intelligent interaction conditions include at least one of the following sub-conditions:

[0056] The calling number is an unfamiliar number;

[0057] The calling number is an unfamiliar number and has no number tag;

[0058] The calling number is an unfamiliar number and has a designated first number mark for artificial intelligence voice interaction;

[0059] The calling number is an unfamiliar number and the calling user identity or calling purpose indicated by the corresponding additional call information requires artificial intelligence voice interaction. The additional call information includes the calling user's identity information and / or call description information;

[0060] The calling number is an unfamiliar number and its number category falls under the preset situation where artificial intelligence voice interaction is required;

[0061] The incoming call time falls within the preset intelligent interaction period;

[0062] The AI ​​voice interaction is manually triggered by the called user.

[0063] Among them, an unfamiliar number refers to a number that is not in the called user's communication contact list. Correspondingly, a known number refers to a number in the called user's communication contact list. Considering that most harassing calls are unfamiliar numbers, artificial intelligence voice interaction can be triggered for all unfamiliar numbers, thereby helping the called user to avoid the impact of harassing calls to the greatest extent.

[0064] Number tag refers to the tag information added by the user or platform to the calling number, such as harassing calls, marketing calls, express calls, takeaway calls, etc. In the prior art, the number tag is stored in the number tag database. Whenever there is an incoming call, the called terminal can query the number tag corresponding to the calling number from the number tag database and display it, so that the called user can understand the nature of the call and make a choice to answer it. In the embodiment of the present application, the number tag can be used as one of the triggering conditions for artificial intelligence voice interaction. For calls with number tags, differentiated processing can be performed based on their number tags, such as directly intercepting fraudulent calls and directly triggering call reminders for takeaway calls. For calls without number tags, there is no information for answering decisions. Therefore, it is necessary to trigger artificial intelligence voice interaction to analyze whether it is a harassing call. Taking the calling number as an unfamiliar number and the absence of a number tag as the intelligent interaction condition can trigger artificial intelligence voice interaction for completely unknown calls, avoiding harassment from such calls. However, the method of intercepting calls based on number tags in the prior art cannot intercept such untagged calls.

[0065] In addition, artificial intelligence voice interaction can be performed on designated number tags. For example, the number tag "banking and financial management" can be set as the first number tag. The first number tag is not limited to one, and it can include multiple different number tags to meet the user's personalized and free customization needs. Then, when the incoming call has the first number tag, artificial intelligence voice interaction is directly triggered to improve the efficiency of call processing.

[0066] Additional call information refers to information sent by the calling user to the called user via the calling terminal, and / or information about the calling user requested by the called terminal from the service end, including but not limited to the calling user's identity information (such as name, company name, etc.) and call description information entered by the calling user (such as the purpose of the call). For example, when a calling user wishes to contact the called user by phone, they can enter some additional information on the calling terminal, such as selecting an electronic business card for the call or entering the purpose or description of the call (such as "discussion about project X"). This information will be sent to the called terminal as part of the additional call information, along with the call request or asynchronously with the call request (the call request is transmitted via a telephone communication link, and the additional call information is transmitted via an Internet link), or uploaded separately to the server. The called terminal can directly receive the additional call information or pull the corresponding additional call information from the server after receiving the call request. The additional call information is then identified based on natural language processing (NLP) technology and machine learning models to determine the caller identity and / or call purpose based on the additional call information. Based on the determined caller identity and / or call purpose, the call is judged whether the call falls under the preset scenario requiring artificial intelligence voice interaction, and if so, the artificial intelligence voice interaction is triggered. In addition, the called terminal can also determine whether the caller identity and / or call purpose identified by the additional call information directly triggers a paid answering process, directly triggers a call reminder, directly triggers a call interception, etc., or make a comprehensive judgment based on other factors, and handle the call according to the corresponding call handling method.

[0067] Number categories refer to information obtained by categorizing phone numbers based on the calling number's call purpose, service category, or the identity of the number holder (i.e., the calling user). These categories are used to characterize the calling number's call purpose, service category, or the number holder's identity. These categories include, but are not limited to, courier calls, marketing calls, food delivery calls, fraudulent calls, official calls, and unclassified calls. A courier call may refer to a call from a courier company or courier, a marketing call may refer to a call from a marketing company or its personnel, a food delivery call may refer to a call from a food delivery platform or a delivery driver, a fraudulent call may refer to a call suspected of being fraudulent, an official call may refer to the office telephone numbers of public institutions such as government agencies, public services, public security, police, and the judiciary, and their personnel, and an unclassified call may refer to a phone number without number identification, registration, or authentication information. Furthermore, the above number categories may also be the result of multi-level classification. Specifically, multi-level classification not only allows for preliminary classification based on the nature and characteristics of the number (such as the call purpose, service category, or the number holder's identity), but also allows for further refinement to meet the personalized needs of users. For example, users may wish to handle marketing calls of different categories differently. Through multi-level classification, marketing calls can be further divided into subcategories such as education and training, financial management, real estate agencies, insurance sales, e-commerce promotions, renovation companies, and recruitment headhunters. These subcategories can even be further subdivided, providing users with a richer and more detailed selection of number categories to choose from. This allows users to decide whether to trigger AI voice interaction or paid answering processes for more specific number categories. This allows users to set more specific and diverse intelligent interaction conditions based on their interests and needs. By classifying number categories into multiple levels and applying them to the setting of intelligent interaction conditions, users can flexibly set multi-level intelligent interaction conditions based on their actual situation and needs, thereby achieving more accurate call management. It should be noted that although the number category and number tag look similar, they are different ways of distinguishing the calling number from different dimensions, and their data sources are different. The number tag is a tag input by the user or the platform for the number, while the number category is obtained by classification based on the known information related to the calling number. For example, the number category can be obtained after analyzing and summarizing at least one piece of information such as the number tag, additional information of the incoming call, and the authentication information of the calling user on the server. The embodiment of the present application supports users to customize multiple number categories for triggering artificial intelligence voice interaction. When an incoming call is detected and the number category of the calling number is determined, it can be matched with the number category customized by the user for triggering artificial intelligence voice interaction. If the match is successful, it is judged that the calling number is an unfamiliar number and its number category belongs to the preset situation where artificial intelligence voice interaction is required, and artificial intelligence voice interaction can be triggered. This embodiment uses number category as one of the intelligent interaction conditions, which has the advantage of high judgment accuracy and can effectively reduce situations such as misjudgment and missed judgment.

[0068] The smart interaction period is a user-defined time period during which artificial intelligence voice interaction is allowed, such as working hours. If it is not within this time period, artificial intelligence voice interaction will not be triggered. Users can decide whether to enable and set this option. If the user sets the smart interaction period, then after detecting an incoming call, the call time can be matched with the smart interaction period. If the match fails, the artificial intelligence voice interaction will not be triggered. If the match is successful, it is necessary to combine other set trigger conditions to comprehensively determine whether to trigger the artificial intelligence voice interaction. For example, if the artificial intelligence voice interaction is set to be triggered for all unfamiliar numbers, then the artificial intelligence voice interaction can be triggered if both the smart interaction period and the unfamiliar number are met. By setting the smart interaction period from the time dimension, the freedom and flexibility of the smart interaction conditions can be improved, making it easier for users to customize smart interaction conditions that better meet their needs.

[0069] There is another scenario where the called terminal can make a pre-alert based on preset conditions after detecting an incoming call, and the user can manually trigger the artificial intelligence voice interaction in the pre-alert interface, thereby meeting the user's demand for free and responsive artificial intelligence voice interaction control based on real-time conditions. In addition, since various mobile phone manufacturers have imposed significant restrictions on the authority to automatically answer incoming calls, it is currently only available for internal use by mobile phone manufacturers. It is difficult for external developers to obtain the authority to automatically answer calls, which affects the promotion and popularization of applications. Therefore, in the case where the authority to automatically answer calls cannot be obtained, the called user can manually trigger the artificial intelligence voice interaction, which can also solve the problem of external developers having difficulty obtaining the authority to automatically answer calls, thereby ensuring the smooth implementation and promotion of this solution, and improving compatibility and adaptability to suit terminals of various brands and models.

[0070] The sub-conditions of the above multiple intelligent interaction conditions can be used one by one, or multiple conditions can be selected and used in combination. When used in combination, any sub-condition can be met to determine that the intelligent interaction condition is met, or all sub-conditions can be met to determine that the intelligent interaction condition is met. These can all be flexibly set and adjusted according to user needs, and the embodiments of this application are not limited thereto.

[0071] The above-mentioned intelligent interaction conditions achieve differentiated processing of unfamiliar calls through multi-dimensional call feature analysis, which can protect users from interference from harassing calls while avoiding the accidental interception of known contacts or important calls. The core logic of the intelligent interaction conditions is to establish a dynamic judgment rule library, combining multiple parameters such as caller number attributes, historical number tags, additional call information and user behavior preferences to selectively trigger the interaction process between the artificial intelligence voice assistant and the calling user. This application is based on the solution of selectively triggering artificial intelligence voice interaction for incoming calls based on intelligent interaction conditions. It can preliminarily judge the intention of the call, which can not only improve the accuracy of handling calls, but also ensure the convenience and personalization of the user experience.

[0072] Regarding how to determine whether the incoming call is a harassing call based on the artificial intelligence voice interaction content, in some embodiments, determining whether the incoming call is a harassing call based on the artificial intelligence voice interaction content includes:

[0073] Determining the calling intention of the calling user based on the artificial intelligence voice interaction content;

[0074] If the incoming call intention meets the first harassing call determination condition, the incoming call is determined to be a harassing call, and the first harassing call determination condition is used to determine whether the incoming call is a harassing call based on the incoming call intention.

[0075] Among them, artificial intelligence voice interaction content refers to the voice content formed by the artificial intelligence voice interaction between the artificial intelligence voice assistant and the calling user. Generally, it is necessary to convert the interactive content into text format through voice-to-text technology for subsequent processing. The interactive content can be analyzed by natural language processing technology (NLP) to extract the calling user's calling intention, such as identifying specific purposes such as "insurance sales". The extracted calling intention can be standardized to facilitate subsequent matching with the intention of harassing calls. Among them, natural language processing technology is an existing technology and only needs to be adaptively adjusted according to the solution of this application to be used in the embodiment of this application to achieve the purpose of this application. Therefore, the details are not repeated here. The first harassing call determination condition can be a standard pre-set according to the call intention for determining whether the call is a harassing call. It can include a set of harassing call intentions obtained by pre-classifying the call intentions of harassing calls. For example, if the calling user's calling intention belongs to the set of harassing call intentions, it is judged that the first harassing call determination condition is met and the call is determined to be a harassing call.

[0076] This implementation method analyzes the calling user's calling intention and combines it with preset judgment conditions to accurately identify harassing calls to trigger the paid answering process. Through analysis, it can deeply understand the real-time intention of the calling user, thereby more accurately distinguishing between legitimate needs and malicious harassment.

[0077] It should be noted that determining the calling intention of the calling user based on the content of artificial intelligence voice interaction is not limited to identifying the calling intention based on the content of artificial intelligence voice interaction. In specific implementation, it can also be combined with other data for comprehensive judgment, such as number characteristics, additional call information, calling behavior data of the calling number, complaint information of the calling number, etc., to identify the calling intention more comprehensively and accurately.

[0078] In some modified implementations, determining the calling user's calling intention based on the artificial intelligence voice interaction content may include:

[0079] Perform semantic recognition on the artificial intelligence voice interaction content according to the semantic recognition model, and determine the calling user's calling intention based on the semantic recognition result; and / or perform keyword matching on the artificial intelligence voice interaction content according to preset calling intention keywords, and determine the calling user's calling intention based on the matching result.

[0080] This embodiment provides two technical paths for identifying the calling user's calling intention: intelligent analysis based on semantic recognition model and matching mechanism based on preset keywords. These two methods can complement each other to improve the accuracy and flexibility of intent recognition and adapt to the needs of different scenarios.

[0081] The semantic recognition model refers to a deep learning model built using natural language processing (NLP) technology, such as pre-trained models like BERT or GPT based on the Transformer architecture. It generates structured intent classification results by understanding the contextual semantics, implicit meaning, and potential needs of voice interaction content. In specific implementation, the AI ​​voice interaction content is first converted into text, then input into the semantic recognition model for analysis. The model outputs call intent labels (such as "express delivery," "real estate agency," "children's training promotion," etc.) and confidence scores for subsequent judgment.

[0082] Preset call intent keywords are lists of keywords or phrases maintained by developers, containing words or phrases strongly related to various call intents, such as "loan," "make-up classes," "insurance," and "interest classes." Through exact matching, phrase matching, or broad matching, the AI ​​voice interaction content is detected to determine whether it contains preset keywords. For example, if the caller mentions "applying for a credit card," keyword matching can directly determine that the call intent is "financial business consultation." Furthermore, the keyword library can be combined with dynamic tagging technology to improve matching accuracy based on historical user interactions or real-time data updates.

[0083] The above two methods can be used either one or in combination. For example, for high-risk calls (such as suspected fraud), the semantic model is preferentially used to analyze the implicit call intention, while for standardized service requests (such as "learn roller skating"), the call intention is quickly determined through keyword matching. In addition, two methods can be used to identify the call intention separately and match them with the first harassing call judgment condition respectively. As long as any one of the matches is successful, it can be determined that the call is a harassing call. The specific implementation can be flexibly changed according to the actual situation, and the embodiments of this application are not limited.

[0084] This embodiment can significantly improve the efficiency and adaptability of intent recognition. The semantic recognition model uses deep learning technology to understand complex contexts and implicit needs, and can cope with challenges such as vague expressions, dialects, or colloquial expressions by callers. The keyword matching mechanism uses rule-based design to achieve fast response and low computing resource consumption, which is especially suitable for high-frequency, standardized call intentions. The combination of the two can balance accuracy and real-time performance. In addition, this embodiment supports dynamic optimization: the semantic model can adapt to emerging intent types (such as new fraudulent techniques) through continuous training, and the keyword library can be expanded independently by administrators or users, such as adding vocabulary in emerging fields. Through this dual-track mechanism, the diversity of call intentions can be more comprehensively covered, reducing the risk of misjudgment or missed judgment, thereby improving user experience and security.

[0085] Regarding how to determine whether the incoming call is a harassing call based on the artificial intelligence voice interaction content, in other embodiments, the above-mentioned determination of whether the incoming call is a harassing call based on the artificial intelligence voice interaction content may include:

[0086] Perform keyword matching on the AI ​​voice interaction content according to preset harassing call keywords, and determine whether the incoming call is a harassing call based on the matching results; or input the AI ​​voice interaction content into a pre-trained harassing call judgment model, and determine whether the incoming call is a harassing call based on the output results of the harassing call judgment model.

[0087] Unlike previous approaches that first identify caller intent based on AI voice interaction content and then identify spam calls based on that intent, this implementation directly utilizes AI voice interaction content to identify spam calls. It also offers two technical approaches for identifying spam calls: a matching mechanism based on preset keywords and intelligent analysis based on pre-trained models. These two approaches combine regularity and intelligence to effectively improve the accuracy and adaptability of spam call identification.

[0088] The pre-set keywords for spam calls are a vocabulary maintained by developers or users, containing high-frequency words or phrases strongly associated with harassing behavior, such as "loan," "winning," "transfer," and "account abnormality." Using exact or fuzzy matching techniques, the AI ​​voice interaction content is detected for the presence of these keywords. If so, the call is classified as spam. It should be noted that some spam calls do not simply mention a single keyword, but instead use a combination of multiple keywords to demonstrate their harassing intent. Therefore, specific keyword combination detection rules can be set to perform keyword matching on the AI ​​voice interaction content. When multiple related keywords appear, the call is classified as spam. For example, the simultaneous presence of "winning" and "handling fee" would indicate a spam call. Furthermore, the spam keyword library can incorporate dynamic update mechanisms, such as real-time updates to scam and telemarketing tactics, to adapt to the evolving nature of new spam tactics.

[0089] The pre-trained spam call identification model is a machine learning or deep learning model built based on a large amount of labeled data. For example, it uses algorithms such as BERT and LSTM to classify spam calls by learning the text features (such as semantic patterns and speech structure) of historical spam calls. During implementation, the text of the voice interaction content is input into the spam call identification model, which outputs the classification result (such as "spamming" or "non-spamming") and the confidence level. For example, when the caller claims that "your express delivery was lost due to the epidemic and a refund is required", the model identifies the combination of keywords such as "epidemic" and "refund" through semantic analysis, and combines the context to determine whether it meets the characteristics of fraudulent speech, thus determining it as a spam call. In addition, the spam call identification model can also be continuously optimized through transfer learning or federated learning to improve its ability to recognize new types of spam patterns.

[0090] The above two methods can be used either one or in combination. For example, high-risk keywords can be directly triggered to intercept, while calls with ambiguous intentions can be deeply analyzed through models. Two methods can also be used to identify harassing calls separately. As long as any method identifies the call as a harassing call, it can be determined that the call is a harassing call. The specific implementation can be flexibly changed according to the actual situation, and the embodiments of this application are not limited.

[0091] This implementation significantly improves the comprehensiveness and efficiency of spam call detection through a combination of rules and intelligence. The keyword matching mechanism quickly identifies high-frequency risk words, enabling immediate interception of obvious spam calls. For example, calls containing clearly labeled spam terms like "winning a prize" and "loan" reduce computing resource consumption and user disruption. The pre-trained model, leveraging deep learning technology to understand complex semantics and underlying intent, can identify implicit fraudulent language (such as requests for "account verification" disguised as "customer service"), avoiding misjudgments or missed calls that could be caused by relying solely on keywords. For example, when a caller uses seemingly legitimate phrases like "system upgrade" but actually requests a money transfer, the model can analyze the unusual intent through context, which keyword matching might fail to identify due to a lack of directly related terms. This implementation supports dynamic scalability: the keyword library can be updated by users or administrators based on the latest fraud cases and spam language, while the pre-trained model can continuously adapt to new spam patterns (such as the characteristics of AI-generated speech). Combining these two mechanisms into a dual-track mechanism provides more comprehensive coverage of the diversity of spam calls, ensuring rapid response to high-frequency risks while addressing complex scenarios through intelligent analysis, significantly improving user experience and communication security.

[0092] It should be noted that the semantic recognition model, harassing call judgment model, etc. involved in the above examples can all be implemented using existing machine learning or deep learning models based on natural language processing (NLP) technology, such as models based on recurrent neural networks (RNN), long short-term memory networks (LSTM), gated recurrent units (GRU), convolutional neural networks (CNN) and Transformer architectures. The embodiments of this application are not limited to this. The above models are all existing technologies and only need to be adaptively adjusted and trained according to the scheme of this application to be used in the embodiments of this application to achieve the purpose of this application. Therefore, the details will not be repeated.

[0093] In some implementations, before step S101, the following steps may also be included:

[0094] Detecting whether the incoming call meets the second harassing call determination criteria;

[0095] If it is detected that the incoming call meets the second harassing call determination condition, triggering the paid answering process, wherein the second harassing call determination condition is used to determine whether the incoming call is a harassing call based on the incoming call information;

[0096] When it is detected that the incoming call does not meet the second harassing call determination condition, a step of detecting whether the incoming call meets the preset intelligent interaction condition is triggered.

[0097] This implementation is a supplementary solution proposed to improve call processing efficiency and meet diverse user needs. It aims to determine whether an incoming call is a spam call based on the caller information before step S101. If it is a spam call, the paid answering process is directly triggered without the need for AI voice interaction or other processes. The caller information can include information such as number characteristics, number tags, additional caller information, and call time that can be obtained without AI voice interaction. Based on this information, it can quickly and directly determine whether an incoming call is a spam call. The second spam call determination condition is a determination rule or model constructed based on the caller information, used to identify calls with high-risk spam characteristics.

[0098] In some implementation variations, the second harassing call determination condition includes at least one of the following sub-conditions:

[0099] The calling number is an unfamiliar number and there is a second number designated for paid answering;

[0100] The calling number is an unfamiliar number and the calling user identity or calling purpose indicated by the corresponding additional call information requires paid answering. The additional call information includes the calling user identity information and / or call description information;

[0101] The calling number is an unfamiliar number and its number category is preset to trigger paid answering directly;

[0102] The call time falls within the preset paid answering time period;

[0103] The calling number has a paid answering indicator.

[0104] Among them, the second number mark is a number mark set by the user to directly trigger the paid answering process. For example, the number mark "loan", "fraud", etc. can be set as the second number mark, so as to directly trigger the paid answering process for incoming calls with the second number mark. It should be noted that the second number mark is not limited to one, and it can also include multiple different number marks, which is not limited in the embodiments of this application.

[0105] In addition, you can also set the triggering conditions for the paid answering process based on the calling user identity or the purpose of the call indicated by the additional information of the incoming call. According to the calling user identity or the purpose of the call indicated by the additional information of the incoming call, it is determined whether the incoming call belongs to the preset situation where the paid answering process needs to be directly triggered. If it belongs to the situation where the paid answering process needs to be carried out, the paid answering process is directly triggered.

[0106] In addition, you can also set the trigger conditions for the paid answering process for the number category, and then determine whether the incoming call belongs to the preset situation where the paid answering process needs to be directly triggered based on the number category. If it belongs to the situation where the paid answering process needs to be carried out, the paid answering process will be directly triggered.

[0107] In addition, the paid answering period is a user-defined time period that can directly trigger paid answering. If it is not within this time period, the paid answering process will not be directly triggered. The user can decide whether to enable and set this option. If the user sets the paid answering period, then after detecting an incoming call, the call time can be matched with the paid answering period. If the match fails, the paid answering process will not be directly triggered. If the match is successful, it is necessary to combine other set trigger conditions to make a comprehensive judgment on whether to directly trigger the paid answering process. For example, if the paid answering process is directly triggered for an unfamiliar number with a second number mark, then the paid answering process can be directly triggered if it meets the paid answering period and the unfamiliar number with a second number mark. By setting the paid answering period from the time dimension, the freedom and flexibility of the second harassing call judgment conditions can be improved, making it easier for users to customize direct triggering conditions for the paid answering process that better meet their needs.

[0108] In addition, the paid answering flag is a user-defined identifier for a known number. It can have two states: true ("true") and null ("null"). This flag allows users to determine whether to enable paid answering for a specific phone number. When the paid answering flag is true, paid answering is triggered for that number. When the paid answering flag is null, paid answering is not. This flag allows users to customize which contacts or phone numbers require paid answering for incoming calls, providing a more personalized and controllable communication management approach. Users can set a paid answering flag for a specific known number through the call application or related service interface. When a known number is set with a paid answering flag by the called user (the paid answering flag changes from an empty value to a true value), and the second harassing call judgment condition is also set to directly trigger the paid answering process for calls with a paid answering flag of a true value, subsequent calls from the known number will directly trigger the paid answering process, thereby helping the called user reduce harassing calls from known numbers or the negative emotions caused by harassing calls. For example, the called user can set this flag to directly trigger paid answering for certain known numbers that frequently disturb but are inconvenient to completely block (such as specific sales calls).

[0109] It should be noted that the above description of each sub-condition in the second harassing call determination condition can be understood and set with reference to the relevant description of each sub-condition in the aforementioned intelligent interaction condition, and some of the content will not be repeated. The above multiple sub-conditions of the second harassing call determination condition can be used one by one, or multiple combinations can be selected. When used in combination, it can be determined that any sub-condition is met to determine that the second harassing call determination condition is met, or it can be determined that all sub-conditions are met to determine that the second harassing call determination condition is met. These can be flexibly set and adjusted according to user needs, and the embodiments of this application are not limited.

[0110] In the embodiment of the present application, the core logic of the second harassing call determination condition lies in establishing a dynamic determination rule base, combining multiple parameters such as caller number attributes, historical number marks, additional call information and user behavior preferences, to selectively and directly trigger the paid answering process. Through multi-dimensional call feature analysis, some calls can be selected to directly trigger the paid answering process, which can eliminate the intelligent interaction process and thus improve the overall call processing efficiency. Moreover, since the second harassing call determination conditions are all user-defined, it can meet the user's needs for flexible, diverse and autonomous direct triggering of the paid answering process according to personal preferences, thereby achieving a more personalized service experience and further improving the user experience.

[0111] In addition, the judgment step of the second harassing call determination condition and the judgment step of the intelligent interaction condition can be executed successively, or they can be combined and judged together. The corresponding process is executed when the condition is met. This embodiment of the present application does not limit this.

[0112] Regarding the above-mentioned paid answering indicator, in some modified implementations, the method may further include:

[0113] In the contact editing interface for the calling number, the paid call indicator setting option is displayed;

[0114] According to the called user's selection operation of the paid answering mark setting option, a paid answering mark is set for the calling number, and the paid answering mark is used to mark that the calling number meets the second harassing call determination condition.

[0115] For example, after the called user answers a paid call from an unknown caller, he or she can choose to save the caller number to the contact list. For example, the called terminal can display a number addition prompt message on the call end interface. The number addition prompt message is used to prompt the called user to add the caller number as a contact, and provide a control (such as a button) for the user to operate (such as clicking the button) to confirm the addition of the caller number. When the called user inputs a contact addition operation indicating that he or she agrees to add the caller number as a contact, a contact editing interface for adding the caller number as a contact can be displayed, and a paid answering identification setting option can be displayed in the contact editing interface. The unselected state of the paid answering identification setting option corresponds to an empty value, and the selected state corresponds to a true value. According to the called user's selection operation of the paid answering identification setting option, a paid answering identification for the above-mentioned calling number can be set. During implementation, the paid answering flag can be set to a true value based on the user selecting the paid answering flag setting option, or the paid answering flag can be set to an empty value based on the user not selecting the paid answering flag setting option. After the contact is saved, the caller number becomes a known number. If the caller number receives another call later, the paid answering flag of the caller number can be used to determine whether the caller number meets the second harassing call determination condition and trigger the paid answering process.

[0116] This implementation provides a convenient solution for setting a paid call identification mark, which can improve the flexibility and accuracy of harassing call management by combining user behavior data with autonomous decision-making. It provides a paid call identification setting option in the contact editing interface, which can convert the user's subjective judgment of the answered number (such as "this number is a sales promotion or fraud") into a system-level defense strategy, avoiding possible misjudgments caused by relying on artificial intelligence.

[0117] In addition, in some modified implementations, the payment confirmation information includes confirmation information that the calling user pays the called user a call fee, and the call fee is determined according to a charging standard preset by the called terminal. The method may further include:

[0118] In the charging standard setting interface, the charging standard editing elements are displayed;

[0119] The charging standard is determined according to the editing operation of the charging standard editing element by the called user.

[0120] The "charging standard" refers to the specific charging standard for paid incoming calls, which is the charging standard set by the called user for incoming calls from the calling number. It can be set by default or flexibly adjusted by the user based on personal needs. It can be a fixed amount for each call, or a dynamic amount based on different call duration thresholds, number categories, and effective time periods. A fixed amount means that all eligible calling numbers are required to pay a fixed amount, while a dynamic amount sets different payment amounts based on different criteria, such as a higher amount for certain types of unfamiliar numbers. For example, a user can set a rule that all unfamiliar numbers are charged a fixed amount of 1 yuan per call, or that certain types of unfamiliar numbers (such as marketing calls) are charged 2 yuan per call. Setting differentiated charging standards helps users manage incoming calls more flexibly based on their needs and preferences. Considering the convenience of payment and the widespread demand for pay-first-then-call models, some embodiments of the present application may adopt a pay-per-call model. This eliminates the need to calculate call charges based on factors such as call duration, reduces system load, improves payment efficiency, and meets the special needs of pay-first-then-remind models.

[0121] The charging standard editing operation can be performed based on the charging standard setting interface. The charging standard setting interface is an interface that can be used to set charging standards, but is not limited to setting charging standards. Charging standard editing elements are controls or components that can be edited and set in the interface, such as input boxes, drop-down menus, sliders, etc. Those skilled in the art can flexibly configure their specific implementation based on actual needs.

[0122] It should be noted that different caller numbers can correspond to the same charging standards or different charging standards. For example, the charging standard for unclassified calls is 1 yuan / time, the charging standard for marketing calls is 2 yuan / time, and the charging standard for fraud calls is 5 yuan / time, etc., thereby raising the threshold for answering various harassing calls to varying degrees and bringing different degrees of benefits to the called users. This graded charging standard helps the called users to set different charging standards for different number categories based on factors such as the answering risks corresponding to various number categories, the intensity of the calling user's call willingness, and the called user's acceptance (or resistance) to the number category. For example, unclassified calls usually refer to those that have not been clearly For unknown numbers that have not been clearly classified, there may be a certain risk of harassment, but the overall risk is relatively low. Therefore, setting a charging standard of 1 yuan per time can not only play a certain screening role, but also will not excessively increase the burden on the calling user; marketing calls usually come from commercial organizations. These calls are often sales-oriented in nature and have a high risk of harassment. In order to reduce the interference of such calls, a higher charging standard of 2 yuan per time can be set, thereby raising the threshold for answering marketing calls; fraud calls are the most risky type of calls. These calls may not only cause property losses, but also threaten the safety of users. Therefore, setting a maximum charging standard of 5 yuan per time can greatly increase the threshold for answering fraud calls and effectively reduce the interference of fraud calls.

[0123] The above charging method in which different number categories correspond to different charging standards can be called a graded charging method. In addition, the graded charging method can also include different charging standards corresponding to different number familiarity levels, different charging standards corresponding to different number sources, different charging standards corresponding to different custom number lists, and different charging standards corresponding to different effective times. Based on the above description, in some changed implementation methods, the charging standards include graded charging standards, and the graded charging standards classify the charging standards from at least one of the following dimensions: number familiarity, number category, custom number list, number source, paid answering mark, membership level and effective time.

[0124] Number familiarity refers to the classification information obtained by categorizing the calling number based on whether it is in the called user's contact list. This information indicates whether the called user recognizes the calling number or the calling user, including both unknown and known numbers. An unknown number refers to a number not in the called user's contact list, while a known number refers to a number in the called user's contact list. A custom number list refers to a user-defined number list. Number origin refers to the number's attribution information based on the calling number's geographic location or carrier information, such as specific region, specific carrier, and international number. A specific region refers to numbers from a specific city or province, a specific carrier refers to numbers from a specific telecommunications company, and international numbers refer to numbers from other countries or regions. Membership tiers refer to the categorization and rating of users in paid answering services based on their membership purchase history, historical behavior, spending history, or other relevant indicators. By setting different membership tiers, users can be provided with a more personalized and differentiated service experience, while also introducing a new dimension to call interactions between callers and called users. The effective time can differentiate charges from a time dimension, such as using higher charging standards during working hours and lower charging standards during non-working hours.

[0125] The tiered charging method can be graded based on at least one of multiple dimensions such as number familiarity, number category, custom number list, number source, paid answering mark, membership level, effective time, etc., where each dimension can correspond to different rate rules. For example, grading based on number familiarity can mean that unfamiliar numbers and known numbers have different charging standards; grading based on number category can mean that different types of phone numbers (such as local numbers, long-distance numbers, international numbers) have different charging standards; grading based on custom number list can mean that numbers on different custom number lists have different charging standards; grading based on number source can mean that numbers from different channels or platforms (such as ordinary telephone networks, VoIP services, etc.) have different charging standards; grading based on paid answering mark can mean that different paid answering marks have different charging standards; grading based on membership level can mean that users of different membership levels (such as ordinary members, senior members, etc.) have different charging standards; grading based on effective time can mean that calls in different time periods (such as working days, non-working days, peak hours, non-peak hours) have different charging standards.

[0126] Determining charging standards based on a tiered charging approach can more accurately manage different types of incoming calls and ensure that called users can flexibly set charging standards based on their own needs and preferences. This flexibility can not only improve users' communication experience, but also effectively balance the interests of calling and called users, ensuring a fair and efficient communication process. In addition, determining charging standards based on a tiered charging approach can also encourage calling users to carefully consider the necessity and urgency of their calls before calling the called users. Higher charging standards can prompt calling users to be more cautious when making high-risk or low-acceptance calls, thereby reducing unnecessary call requests and improving the quality and efficiency of calls. In short, through this multi-level, differentiated charging standard, the needs of different users can be better met and the quality of the overall communication environment can be improved.

[0127] In some embodiments, after step S102, the following may also be included: according to the call handling method set by the user for non-harassing unfamiliar calls, a regular answering process is executed for the unfamiliar calls that are judged not to be harassing calls, wherein the regular answering process is a call handling process executed according to the call handling method set by the user for non-harassing unfamiliar calls, and may be any answering process other than a paid answering process, including but not limited to triggering a call reminder, directly intercepting blacklisted numbers, using an artificial intelligence voice assistant to answer normal calls, forwarding calls to voicemail, etc., which are not limited in the embodiments of this application. For example, considering that the calls that are judged not to be harassing calls after step S102 are still unfamiliar numbers, the user can set an appropriate call handling method for the unfamiliar numbers that are non-harassing calls, such as directly triggering a call reminder, using an artificial intelligence voice assistant to answer normal calls, etc., thereby ensuring that non-harassing unfamiliar calls are answered normally.

[0128] In other embodiments, after step S102, the method may further include: according to the incoming call intention determined based on the artificial intelligence voice interaction content, selecting an incoming call processing method corresponding to the incoming call intention to handle the incoming call.

[0129] In this embodiment, corresponding call handling methods can be pre-set for various call intentions. The call handling methods include but are not limited to triggering call reminders, directly intercepting calls, answering calls through artificial intelligence voice assistants, forwarding calls to voicemail, triggering paid answering processes, etc. The call handling methods corresponding to different call intentions can be the same or different. For example, some call intentions correspond to the same call handling method, and other call intentions correspond to another call handling method. For example, in the case of call intentions for express delivery or takeout, an artificial intelligence voice assistant can be used to answer the call, and the caller can be informed of the delivery method and location of the express delivery and takeout according to the preset script. In the case of call intentions for customer calls, a call reminder can be triggered to remind the called user to answer the call. Through this embodiment, it is possible to select an appropriate answering method for non-harassing unfamiliar calls based on the content of the artificial intelligence voice interaction, further improving the user experience.

[0130] Taking into account that if the incoming call rings immediately, it will inevitably cause harassment to the called user, therefore, in some embodiments, answering the incoming call and performing artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant may include: for the incoming call, answering the incoming call and performing artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant while keeping the incoming call silent at the foreground of the called terminal.

[0131] Among them, keeping incoming calls silent at the front desk means that no incoming call reminder will be issued on the called terminal, that is, neither the incoming call reminder will be displayed nor the ringtone or vibration will be played, so as to avoid the incoming call reminder being issued before the calling user confirms the payment and causing harassment to the called user. The incoming call silence will not be lifted and the incoming call reminder will be triggered until the payment confirmation information corresponding to the calling number is detected. This can ensure that the called user will answer the call for which he is willing to pay after receiving the call reminder, thereby avoiding interference to the called user by calls that are unwilling to pay.

[0132] Considering that the called user may not answer the call in time, if the call reminder is triggered only after the payment confirmation information is detected, the calling user may have to wait for a long time before the call is answered, or even not be answered; in other cases, the calling user may not be willing to pay the answering fee, but the called user may be willing to answer the call because he is bored or knows who is calling. If the call reminder cannot be triggered because the calling user does not confirm the payment, it is a loss for both the calling and the called party; in other cases, the mobile phone manufacturer restricts the automatic connection permission of incoming calls. The user manually triggers the AI ​​answering or paid answering process; in such situations, the called user needs to be informed of the incoming call information in advance before the call reminder. Therefore, in some modified implementation methods, the method may also include: an incoming call pre-reminder step; the incoming call pre-reminder step includes: for an incoming call, when it is detected that the pre-reminder conditions are currently met, triggering the incoming call pre-reminder, the incoming call pre-reminder includes displaying the incoming call pre-reminder interface, and the incoming call pre-reminder interface includes at least one of an answer trigger control, a reject trigger control, a paid answer trigger control and an AI answer trigger control.

[0133] In addition to displaying the pre-alert interface, the aforementioned pre-alert can also include reminders such as ringing and vibration, forming a secondary reminder mechanism with subsequent call reminders. This increases the probability of the called party answering the call if the calling user is willing to pay for the call. Furthermore, the pre-alert interface includes answer trigger controls, a reject trigger control, a paid answer trigger control, and an AI-powered proxy trigger control, providing users with more comprehensive and flexible call handling options to meet their diverse call handling needs.

[0134] Regarding the execution time of the call pre-alert, it can be executed immediately after receiving the call request, or before detecting whether the call meets the preset intelligent interaction conditions, or before detecting whether the call meets the second harassing call judgment conditions. Those skilled in the art can flexibly set it according to actual needs and the overall solution, and the embodiments of this application are not limited. In addition, in addition to the above-mentioned multiple trigger controls, the content displayed on the call pre-alert interface can also include call information, such as caller number, number tag, number category, call additional information, call intention, etc., to help the called user make a decision to answer the call, and the embodiments of this application are not limited.

[0135] The above-mentioned pre-reminder conditions can be flexibly set according to the actual application scenarios. The current pre-reminder conditions may include at least one of the following: the pre-reminder function is currently turned on, the current scene meets the preset pre-reminder scene, the calling number belongs to the pre-reminder number category, the calling number belongs to the pre-reminder number list, and the current time is in the pre-reminder period.

[0136] The pre-alert feature is a service option. When enabled, it triggers a pre-alert upon detecting a call request. The pre-alert feature can be enabled or disabled by the user. Furthermore, if automatic call connection permission is denied, the pre-alert feature can be forcibly enabled and triggered, allowing the called user to manually trigger AI voice interaction or a paid call answering process.

[0137] Pre-alert scenarios are user-configurable conditions that automatically determine whether to trigger a pre-alert based on the current device environment or usage. For example, the system can determine whether the current scenario meets a pre-alert scenario based on factors such as the called device's location, current application usage, and ambient noise level. If so, the pre-alert will be automatically triggered.

[0138] The pre-alert number category refers to the number category that can trigger the incoming call pre-alert. For the definition and description of the number category, please refer to the aforementioned embodiment description of the number category for understanding. The pre-alert number category can be set by the call application by default, or it can be set and modified by the user, and the embodiments of this application are not limited to this. By setting the pre-alert number category to determine whether an incoming call can trigger the incoming call pre-alert, it can fully meet the user's personalized setting needs and achieve the purpose of pre-alert only for incoming calls of certain number categories.

[0139] The pre-alert number list is a user-defined list of numbers that lists specific phone numbers that the user wants to automatically trigger pre-alerts when a call comes in. By adding specific phone numbers to the pre-alert number list, users can respond quickly when receiving calls from these numbers, ensuring that they do not miss important call opportunities.

[0140] Pre-alert periods are user-defined time periods during which pre-alerts are automatically triggered for all incoming calls that meet the criteria. For example, users can set pre-alert periods during weekday office hours or evening breaks to more flexibly handle incoming calls during these times. By properly configuring pre-alert periods, users can receive and handle important calls promptly without disrupting their daily lives and work, thereby improving communication efficiency and personal satisfaction.

[0141] By introducing pre-alert condition elements such as pre-alert function, pre-alert scenario, pre-alert number category, pre-alert number list, and pre-alert time period, it is possible to provide a more intelligent and personalized communication management solution while protecting user rights, ensuring that every successful call can meet the actual needs of both parties, improving the professionalism and reliability of the service, and enhancing users' trust in paid answering services. For example, a pre-alert function switch can be set in the call application. If the called user turns on the pre-alert function switch, the pre-alert function will take effect. At this time, the incoming call meets the pre-alert conditions and can trigger the pre-alert. For example, further pre-alert scenarios can be set. For example, the pre-alert for an incoming call can be executed when the called terminal screen is on. The on screen indicates that the user is using the terminal, and executing the pre-alert in this case can achieve better results. For example, the pre-alert for an incoming call can be executed in scenarios other than full-screen game running. Since users often do not want to be disturbed when playing full-screen games, executing the pre-alert for an incoming call in scenarios other than full-screen game running can effectively avoid pre-alert interference to users. Those skilled in the art can set more pre-alert scenarios for users to choose from according to actual needs to meet users' needs for more flexible and personalized pre-alert settings.

[0142] It should be noted that the incoming call pre-reminder interface can be a full-screen interface or a partial interface, and the embodiments of the present application do not limit this. For example, the incoming call pre-reminder interface can be displayed in full screen when the terminal is idle to achieve a significant reminder effect. When the terminal is running a full-screen game, video conference, and playing a video, the incoming call pre-reminder interface can be displayed partially, which not only has a reminder effect but also reduces the degree of interference with the user's current work.

[0143] In the case where the called user is willing to answer the call for free because he is bored or knows who is calling, he can directly answer the call based on the pre-call reminder. Accordingly, after the pre-call reminder is triggered, the method may further include: answering the call according to the direct answering operation input by the called user in response to the pre-call reminder. The direct answering operation includes a first operation on the answering trigger control in the pre-call reminder interface, and the first operation is used to trigger the answering of the call. In this way, the pre-call reminder can not only remind the called user that there is a new call, but also provide an answering option for the called user to choose whether he is willing to answer the call for free, and directly answer the call when a direct answering operation is detected, thereby eliminating the artificial intelligence voice interaction process and payment steps, and meeting the user's answering needs in specific scenarios.

[0144] In the case where the called user is busy or temporarily inconvenient to answer the call for other reasons, the calling user does not need to continue to pay, because even after paying, the called user will not answer the call. In this case, the call can be directly rejected based on the call pre-reminder without the need for unnecessary processes such as artificial intelligence voice interaction. Accordingly, after triggering the call pre-reminder, the method may further include: rejecting the call according to the rejection operation input by the called user in response to the call pre-reminder. The rejection operation includes a second operation on the rejection trigger control in the call pre-reminder interface, and the second operation is used to trigger the rejection of the call. In this way, the call pre-reminder can also provide a rejection option for the called user to choose whether to directly reject the call, and directly reject the call when the rejection operation is detected. On the one hand, unnecessary steps such as artificial intelligence voice interaction and payment can be avoided, and on the other hand, the call reminder will not be triggered again to cause subsequent disturbances to the called user.

[0145] In addition, the call pre-reminder also supports the user to manually trigger the paid answering process. Accordingly, after the call pre-reminder is triggered, the method may further include: triggering the paid answering process according to the paid answering operation input by the called user in response to the call pre-reminder; wherein the paid answering operation includes a third operation on the paid answering trigger control in the call pre-reminder interface, and the third operation is used to trigger the paid answering operation. In this way, the call pre-reminder can also provide a paid answering option for the called user to choose whether to directly trigger the paid answering process, and directly trigger the paid answering process for the call when a paid answering operation is detected, thereby omitting the artificial intelligence voice interaction process, improving the overall process efficiency, and providing a new paid answering plan triggering entry in the absence of automatic call connection authority to ensure the smooth implementation of the plan.

[0146] In addition, the call pre-reminder also supports users to manually trigger artificial intelligence voice interaction. Accordingly, after the call pre-reminder is triggered, the method may further include: according to the artificial intelligence answering operation input by the called user in response to the call pre-reminder, triggering the step of answering the call and conducting artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant; wherein, the artificial intelligence answering operation includes a fourth operation on the artificial intelligence answering triggering control in the call pre-reminder interface, and the fourth operation is used to trigger the artificial intelligence answering operation. In this way, the call pre-reminder can also provide an artificial intelligence answering option for the called user to choose whether to directly trigger the artificial intelligence voice interaction step, and directly trigger the artificial intelligence voice interaction step for the call when the artificial intelligence answering operation is detected, thereby providing a new artificial intelligence voice answering triggering entry in the absence of automatic call connection permission, solving the problem of dependence on automatic call connection permission, and ensuring the smooth implementation of the solution.

[0147] In addition, if no operation is detected within the preset time, that is, no operation is input for the incoming call pre-reminder, the incoming call pre-reminder can be automatically closed and the subsequent process can continue, such as detecting whether the incoming call meets the preset intelligent interaction conditions or detecting whether the incoming call meets the second harassing call judgment conditions, etc. The above-mentioned preset time can be 3 seconds, 5 seconds, 10 seconds, etc., and can be flexibly set according to actual needs. The embodiment of this application is not limited. Its value is intended to be able to play a pre-reminder role without disturbing the user for too long, so it should be a relatively short time.

[0148] It should be noted that the above-mentioned answer trigger control, reject trigger control, paid answer trigger control and artificial intelligence agent trigger control can be different controls or the same control. For example, the answer trigger control, reject trigger control, paid answer trigger control and artificial intelligence agent trigger control can be four virtual buttons, and the first operation, the second operation, the third operation and the fourth operation are respectively click operations or drag operations on the above-mentioned answer trigger control, reject trigger control, paid answer trigger control and artificial intelligence agent trigger control; for another example, the answer trigger control, reject trigger control, paid answer trigger control and artificial intelligence agent trigger control can be the same slider, and the first operation, the second operation, the third operation and the fourth operation are respectively dragging the above-mentioned slider to the right or left. Operations in the same position or direction; for example, the answering trigger control, the rejecting trigger control, the paid answering trigger control and the artificial intelligence proxy trigger control can be four area trigger controls, and the first operation, the second operation, the third operation and the fourth operation are respectively dragging another trigger control to the range of the above four area trigger controls, etc.; the above methods all have the advantages of being simple and easy to operate. In addition, the direct answering operation is not limited to the above first operation, the rejecting operation is not limited to the above second operation, the paid answering operation is not limited to the above third operation, and the artificial intelligence proxy answering operation is not limited to the above fourth operation. They can also be implemented through gesture operations, voice control operations, etc. The embodiments of this application are not described one by one, and they should all be within the scope of protection of this application.

[0149] The embodiment of the present application, by introducing a pre-call reminder mechanism, can provide a more intelligent and personalized communication management solution while protecting the rights and interests of users, especially providing more flexible and personalized services in different application scenarios. It can not only simplify the process of establishing calls and improve communication efficiency, but also enhance user satisfaction with the paid answering mechanism. For example, a user receives a pre-call reminder that says "You have a call from an unfamiliar number of a decoration company. Please choose to answer for free, answer for a fee, reject or answer on behalf of AI." If the user is preparing for decoration and is currently bored, he can answer the call directly; if the user is busy at work or it is inconvenient to answer the call, he can choose to reject the call; if the user has no decoration needs but is currently bored, he can choose to answer for a fee or answer on behalf of AI and wait for the calling user to pay before answering the call.

[0150] It should be noted that the reminder methods of the incoming call pre-alert and the incoming call reminder can be the same or different, and this embodiment of the application does not limit this. Considering that if the calling user is unwilling to pay the answering fee, the above-mentioned incoming call pre-alert will also cause interference to the called user to some extent, therefore, in some embodiments, the above-mentioned incoming call pre-alert can be implemented using a method with a lower reminder intensity than the incoming call reminder. Reminder intensity refers to the effect and strength of the reminder method in attracting the user's attention. Different reminder methods have different abilities to attract the user's attention, and the strength of this ability is the reminder intensity. For example, on the basis that both the incoming call reminder mode and the incoming call pre-reminder mode have screen prompts, the incoming call reminder mode and the incoming call pre-reminder mode also include ringing, but the ringing volume of the incoming call pre-reminder is smaller than the ringing volume of the incoming call reminder, so that the reminder intensity of the incoming call pre-reminder is smaller than the reminder intensity of the incoming call reminder; for another example, the incoming call reminder mode also includes ringing, and the incoming call pre-reminder mode also includes vibration, so that the reminder intensity of the incoming call pre-reminder is also smaller than the reminder intensity of the incoming call reminder; for another example, the incoming call reminder mode and the incoming call pre-reminder mode also include vibration, but the vibration duration of the incoming call pre-reminder is smaller than the vibration duration of the incoming call reminder, so that the reminder intensity of the incoming call pre-reminder is also smaller than the reminder intensity of the incoming call reminder. By limiting the reminder intensity of the incoming call pre-reminder to be smaller than the reminder intensity of the incoming call reminder, it can not only play the role of pre-reminder, ensuring that the called user is informed of the incoming call information in advance before the incoming call reminder, thereby increasing the probability of the incoming call being answered, and meeting the called user's diverse call processing needs, but also can greatly reduce the interference of the pre-reminder on the user.

[0151] In some example embodiments, considering that whether the calling number triggers the paid answering process is determined by the called terminal, the called terminal should send a paid answering prompt message to the calling terminal to notify the calling user to pay the answering fee. If the calling user agrees to pay, it is also necessary to send payment instructions to the calling terminal so that the calling user can pay the answering fee according to the payment instructions. In addition, it may be necessary to send payment timing option information to the calling terminal for the calling user to select the payment time, including payment before answering or payment after answering, etc. Therefore, in some modified implementations of the embodiments of the present application, the paid answering process also includes a paid answering information interaction step, which includes exchanging paid answering information with the calling user through an artificial intelligence voice assistant, and the paid answering information includes at least one of paid answering prompt information, payment instructions, payment timing option information, and automatic payment option information.

[0152] The paid answering reminder information is information used to remind the calling user that the calling user needs to pay the answering fee, for example, reminding the calling user to pay the answering fee before reminding the called user to answer the call.

[0153] The payment guidance information is information used to guide the calling user to pay the answering fee, and may include payment parameters required for the calling user to pay the answering fee, usually including at least one of the collection account, payment amount, payment method, payment link, and payment order number, to ensure that the calling user can successfully complete the payment operation. Specifically, the collection account may be an account set by the called user for collecting the answering fee, that is, the first account mentioned in the aforementioned embodiment. In the case where there is a payment platform as an intermediate guarantee, the collection account may also be an intermediate account corresponding to the payment platform, which is not limited in the embodiment of the present application; the payment amount is the specific answering fee that the calling user needs to pay; the payment method is a method for the calling user to choose how to pay, which may include a variety of payment options, so that the calling user can choose the most convenient way to pay. Common payment methods include but are not limited to bank transfers, third-party payment platforms (such as Alipay, WeChat Pay, etc.), credit card payments, etc. Each payment method will provide corresponding payment guidance to help the calling user complete the payment operation; the payment link is a link that can be clicked directly to make a payment, which simplifies the payment process. The calling user can click on the link to make a payment. The link jumps directly to the payment page for payment. The payment link is usually generated by the payment platform and transmitted to the calling user through a secure encrypted channel to ensure the security of the payment process. The payment order number is a unique number for the payment order generated for each call. The called terminal or the server can directly or through a third-party payment platform generate a payment order (also called a paid answering order) for each harassing call or harassing call that agrees to pay the answering fee, and send the unique number of the payment order, namely the payment order number (also called the paid answering order number), to the calling terminal. The calling terminal can directly call up the payment order and make payment according to the payment order number. Those skilled in the art can flexibly set the content of the above-mentioned payment guidance information according to the payment environment and actual needs, so as to achieve the purpose of the calling user paying the answering fee to the called user according to the payment guidance information. Optionally, in order to improve security, the complete account information is usually not sent directly, but a one-time payment link, payment order number or other security mechanism is used to protect the account information from being leaked.

[0154] The payment timing option information is information for users to select the payment timing, which includes at least one payment timing option for the calling user to select. The payment timing option includes a post-pay option and / or a pre-pay option. The post-pay option means that the answering fee will be paid after the call ends, and the pre-pay option means that the answering fee will be paid first and then the incoming call reminder is triggered.

[0155] The automatic payment option information is used by the calling user to select whether to use the automatic payment mechanism to pay the call fee. It should be noted that this application does not limit the specific information content of the above-mentioned various paid call information. As long as the meaning expressed in the content is consistent with the above description, it shall be within the scope of protection of this application.

[0156] It should be noted that the above multiple paid answering information can be sent in sequence according to actual needs, or can be sent partially or all at the same time. For example, a paid answering prompt message can be sent to the calling terminal first, and after receiving feedback information that the calling terminal agrees to pay, the payment guidance message can be triggered to be sent to the calling terminal; the payment guidance message can also be sent to the calling terminal together with the paid answering prompt message, so that the calling user can make payment directly if he agrees to pay, which is more efficient and helps to improve the calling user's call efficiency and user experience. Those skilled in the art can make reasonable expansions and changes based on the above examples, which should all be within the scope of protection of this application.

[0157] By exchanging paid answering information with the calling user through the artificial intelligence voice assistant, the efficient transmission of paid answering information can be effectively realized. Regardless of whether a specific application is installed on the calling terminal, the paid answering information can be received and broadcasted, which has high compatibility. In addition, the artificial intelligence voice assistant exchanges paid answering information with the calling user in voice through the call link, which is more intuitive and efficient for the calling user and easy for the calling user to accept. In addition, there is another advantage of exchanging paid answering information with the calling user through the artificial intelligence voice assistant after the call is connected, that is, the telecom operator has set a certain waiting time (also called ringing time or ringing cycle) for the current call technology, which is used to limit the number of calls from the calling party to the phone. When a call starts, the phone starts ringing until the call is answered or the call is automatically hung up due to timeout, such as 30 seconds, 45 seconds, etc. If the call is not answered within the waiting time, a busy tone will be prompted and the call will be automatically hung up. However, the calling user may not have time to complete the payment within the waiting time, resulting in the call being automatically hung up, or the call may be automatically hung up just after the payment is completed. This will seriously affect the calling user's experience and may also cause the called user to miss the call. However, this application uses an artificial intelligence voice assistant to interact with the calling user for paid answering information after the call is answered, which directly skips the above-mentioned waiting time and avoids the call being automatically hung up, thereby ensuring that the calling user's call and payment can be completed smoothly, and avoiding the called user from missing the call.

[0158] It should be noted that the above-mentioned paid answering information interaction steps are not limited to being implemented through the artificial intelligence voice assistant as a paid answering information interaction method. It can also be implemented through or in combination with other paid answering information interaction methods. Other paid answering information interaction methods may include but are not limited to at least one of the following: interacting with the calling user for paid answering information through an interactive voice response system, sending paid answering information to the calling terminal through in-application notifications, and sending paid answering information to the calling terminal through text messages.

[0159] The above-mentioned interactive voice response system can be an external interactive voice response system provided by a telecommunications operator or a third-party service provider, or it can be a built-in interactive voice response system jointly implemented by a first call application and a second call application, both of which can achieve the purpose of the embodiments of the present application. In some embodiments, the sending of paid answering information to the calling terminal through the interactive voice response system includes: sending paid answering information to the calling terminal through the built-in interactive voice response system; wherein, the built-in interactive voice response system is jointly implemented by a first call application installed on the called terminal and a second call application installed on the calling terminal, the first call application sends paid answering information to the second call application through a call link or an Internet link, and the second call application sends response information for the paid answering information to the first call application through an Internet link.

[0160] In-app notification refers to sending notifications or prompt information to the calling terminal through the message push function within the application. For example, the first call application can use in-app notifications to send paid answering information to the second call application, and display it on the display interface of the calling terminal. Since the reading speed of visual information is higher than the listening speed of voice information, sending paid answering information to the calling terminal through in-app notifications is highly intuitive and helps to improve call efficiency and user call experience.

[0161] Sending paid answering information to the calling terminal via SMS is to use the SMS service in the mobile communication network to send a text message containing paid answering information to the calling terminal. After the calling user sees the text message, he can directly click on the answer link in it to answer, or reply to the text message to answer. The called terminal, the first call application or the server can identify the answer information of the calling user based on the answer link clicked by the calling user or the content of the reply text message, and understand his payment intention, etc. Among them, the text message sent to the calling terminal can contain multiple answer links, and different answer links correspond to different answer information. In this way, after the calling user clicks the answer link, the called terminal can determine the answer information of the calling user by detecting the trigger information of each answer link, and then identify the payment intention of the calling user, etc.

[0162] In addition, considering that different calling terminals support different paid answering information interaction methods, in some embodiments, a paid answering information interaction method applicable to the calling terminal can also be selected based on the device information of the calling terminal. The above-mentioned paid answering information interaction step may include: according to the device information of the calling terminal, selecting a paid answering information interaction method applicable to sending the currently to-be-sent paid answering information to the calling terminal from the preset paid answering information interaction methods, the paid answering information interaction method including at least one of interacting with the calling user for paid answering information through an artificial intelligence voice assistant, interacting with the calling user for paid answering information through an interactive voice response system, sending paid answering information to the calling terminal through in-application notifications, and sending paid answering information to the calling terminal through text messages; sending the paid answering information to the calling terminal according to the determined paid answering information interaction method. In this way, it is possible to select a more appropriate paid answering information interaction method to send paid answering information based on the device information of the calling terminal and the currently to-be-sent paid answering information, and adopt the most appropriate method to interact with the calling user, thereby further improving the rationality and efficiency of the paid answering information interaction, as well as improving the acceptance and interaction experience of the calling user, thereby reducing the difficulty of popularizing and promoting the present application scheme and improving the user experience.

[0163] The aforementioned device information refers to hardware and software characteristic data related to the calling terminal, such as device type (landline, mobile phone, virtual phone, etc.), operating system type, version number, whether specific applications (such as second call applications) are installed, etc. This data can help the called terminal or the server understand the capabilities and limitations of the calling terminal, and thus intelligently select the most suitable paid call information interaction method for the calling terminal based on this information. For example, for traditional communication devices such as landline phones, it can be recognized that they do not have intelligent operating systems, application installation capabilities, and online payment functions, and artificial intelligence voice assistants can be used to send paid call information; while for modern mobile devices such as smartphones, more detailed attributes such as their operating system type (such as Android or iOS), version number, screen size, network connection status, and whether specific applications (such as second call applications) are installed can be detected. If a second call application is installed, both artificial intelligence voice assistants and in-app notifications can be selected for paid call interaction.

[0164] In some embodiments, the called terminal can simultaneously support the above multiple paid answering information interaction methods based on the first call application, and first detect the device information and call environment of the calling terminal before the paid answering information interaction, such as whether the second call application is installed, etc., and select the appropriate paid answering information interaction method based on the device information, call environment and pre-set priority of the paid answering information interaction method of the calling terminal to send paid answering information to the calling terminal, so as to interact with the calling user for paid answering information. Those skilled in the art can make reasonable expansions and changes based on the above examples, which should also be within the scope of protection of this application.

[0165] It should be noted that the above-mentioned multiple paid answering information interaction methods can be implemented separately or in combination without conflict. For example, when both the calling terminal and the called terminal are installed with corresponding call applications, the artificial intelligence voice assistant can be used to send paid answering information to the calling terminal, and the in-application notification can be used to send paid answering information to the calling terminal at the same time, thereby improving the delivery rate of paid answering information in a variety of ways; for example, the artificial intelligence voice assistant can be used to interact with paid answering prompt information, and the in-application notification can be used to send payment guidance information, etc.; those skilled in the art can expand the implementation based on this exemplary description, and this embodiment is not limited. In addition, those skilled in the art can also adopt other changes to send paid answering information to the calling terminal under the inspiration of the above-mentioned multiple paid answering information interaction methods. They will not be described one by one here, but they should all be within the scope of protection of this application under the same technical concept.

[0166] It should be noted that after the calling terminal receives and plays or displays the above-mentioned paid answering information, the calling user is generally required to respond, such as confirming whether or not to agree to pay the answering fee, and send the response information to the called terminal so that the called terminal can continue the subsequent process according to the response information, such as continuing the subsequent answering process according to the payment consent information indicating agreement to pay the answering fee, and triggering the rejection of the call according to the payment refusal information indicating disagreement to pay the answering fee. The calling user can trigger the reply information (i.e., the response information) by key reply, in-application confirmation, or voice reply. Key reply refers to the calling user generating a reply message by operating the designated key on the calling terminal. It can reply to the above-mentioned multiple paid answering prompt methods. It is a relatively simple, convenient and highly versatile reply method that does not require additional equipment or software support and is suitable for all types of calling terminals. If both the calling terminal and the called terminal have installed the same call application, then in-application confirmation is also a convenient and secure method. For example, when the calling terminal receives a paid answering prompt message, the call application will display a selection interface on the screen of the calling terminal, prompting the user to choose to agree to pay or refuse to pay, and provide corresponding selection controls. The calling user can choose to agree to pay or refuse to pay through the above selection controls. The call application will generate a payment consent message and send it to the called terminal if the calling user chooses to agree to pay, or generate a payment refusal message and send it to the called terminal if the calling user chooses to refuse to pay. The advantage of this method is that the user experience is good, the interactive interface is intuitive, and more content can be provided in the interface, such as the payment platform interface, to support the calling user to pay directly without selecting the payment consent option, thereby simplifying the overall process. Voice response refers to the calling user expressing his or her willingness to pay through voice response. For example, when using an artificial intelligence voice assistant to send paid answering information, the calling user can express his or her agreement to pay or refusal to pay through voice reply, and the called terminal or server can confirm the calling user's intention through voice recognition. This method is suitable for users who are unable or inconvenient to use button operations, such as when driving or when the hands are inconvenient. It is highly natural and flexible.

[0167] The answering fee paid by the calling user to the called user can be paid directly, for example, the second account held by the calling user pays directly to the first account held by the called user, or it can be paid indirectly, for example, the second account held by the calling user first pays to the intermediate account, and then the intermediate account pays to the first account held by the called user, as long as the called user can ultimately benefit from answering the call. The payment confirmation information can be triggered and generated based on the payment behavior of the calling user, or it can be triggered and generated after the answering fee arrives at the called user's account, that is, the payment confirmation information can be triggered and generated after the calling user completes the payment to the intermediate account, or it can be triggered and generated after the intermediate account completes the payment to the called user. This is not limited in the embodiments of the present application. In addition, the answering fee paid by the calling user to the called user can be paid in full to the called user, or partially to the called user. For example, part of the answering fee paid by the calling user is given to the called user, and the other part is given to the platform account corresponding to the first call application as a technical service fee, etc., which does not harm the purpose of the called user as the beneficiary of the paid answering, is in line with the invention purpose of the embodiments of this application, and should be within the scope of protection of this application.

[0168] It should be noted that the payment confirmation information can be payment completion information, that is, information confirming that the calling user has completed payment, indicating that the answering fee corresponding to the calling number has been paid. The payment completion information can be triggered based on the payment behavior of the calling user, such as after the calling user completes payment to the intermediary account, or after the answering fee reaches the called user's account, such as after the intermediary account completes payment to the called user, thereby ensuring that the called user can receive the answering fee before triggering the call reminder. In addition, the payment confirmation information can also be payment commitment information, that is, information that the calling user confirms that payment will be made, indicating that the calling user promises to pay the answering fee for this call, thereby triggering the call reminder first if the calling user promises to pay, and then making payment after the call is completed. By moving the payment process to the end of the call, the call efficiency can be effectively improved. The above two methods can be used to achieve the purpose of the embodiments of the present application, which are explained below.

[0169] For the first case, the payment confirmation information can be payment completion information. Payment completion information refers to information confirming that the payment has been successfully completed. The payment completion information can be generated by a payment platform (such as a bank, a third-party payment system, etc.) and sent to the server or the first call application of the called terminal through a secure channel. In addition, the first call application or the server can also detect whether the payment is completed by monitoring the active account information of the first account held by the called user, such as querying the payment completion information corresponding to the calling number in the active account information. If it is found, a call reminder can be triggered. In addition, any payment completion information acquisition method provided by the prior art can be used to obtain the payment completion information. These methods will not be described here one by one, and they are all within the scope of protection of this application. The payment completion information may include but is not limited to key data such as payment status, payment amount, payment timestamp, payment transaction ID (i.e., payment order number), and the calling number associated with the payment. When the called terminal obtains and verifies the payment completion information, it confirms that the payment has indeed been completed, and then the call silent state will be released and the call reminder will be triggered. This method can ensure that the called user can make a call after actually receiving the fee, thereby protecting the rights and interests of the called user.

[0170] For the second case, the payment confirmation information can be a commitment payment information. The commitment payment information refers to the information that the calling user confirms that the payment will be made. In this case, the calling user has not actually completed the payment operation, but has clearly expressed his willingness to pay the answering fee. The commitment payment information can be transmitted in a variety of ways, such as through button reply, in-application confirmation or voice answer. After receiving the commitment payment information, the called terminal can temporarily lift the silent state of the incoming call and trigger an incoming call reminder. This method can provide a faster response speed and help improve call efficiency. It is easy to understand that this method is more suitable for user groups with higher trust, or in situations where a call needs to be established quickly in an emergency, but other situations are also applicable.

[0171] Based on the above description of payment confirmation information, the embodiment of the present application supports at least two paid answering modes, one is a pay-first-then-answer mode, in which the payment confirmation information is the payment completion information; the other is an answer-first-then-pay mode, in which the payment confirmation information is the payment commitment information. During implementation, the above paid answering mode can be provided to the called user for selection and setting. The user can choose one of the paid answering modes or choose to accept the above two paid answering modes. Different paid answering modes can also be set for different number categories. In addition, different paid answering modes can also be set according to the caller number's credit information. For example, for lost signal codes and fraudulent calls, only the pay-first-then-answer mode is supported, and for unclassified calls and non-lost signal codes, the above two paid answering modes are supported. After the setting is completed, the paid answering process can be executed according to the paid answering mode selected by the user, and a suitable paid answering mode can be provided for the calling user to select based on the number category and credit information of the calling number.

[0172] In the case where the called user chooses to support the above two paid answering modes, the corresponding payment timing option information can be generated and sent to the called terminal for selection, wherein the pay first and answer later mode corresponds to the pay first option, and the answer first and pay later mode corresponds to the pay later option. The above payment timing option information can be sent separately, for example, after the paid answering prompt information, or it can be sent together with the above paid answering prompt information or payment instruction information, or it can be sent as part of the above paid answering prompt information or payment instruction information, which is not limited in the embodiments of the present application. After the calling terminal plays or displays the above payment timing option information, the calling user can select the payment timing option that suits him / her. If the calling user selects the pay later option, the committed payment information can be triggered, and if the calling user selects the pay first option, it is necessary to make the payment and trigger the payment completion information after the payment is completed. It should be noted that the above-mentioned post-payment option and pre-payment option can be displayed or played together, or one of them can be displayed or played selectively. For example, only the above-mentioned post-payment option can be displayed, or the payment method options (such as bank transfer, third-party payment, in-app wallet payment) can be displayed at the same time. The above-mentioned payment method option can also implicitly indicate the meaning of paying first and then answering the call, and can also be regarded as a pre-payment option, which is just merged with the specific payment method.

[0173] In some examples, after receiving a paid answering prompt message, the calling terminal will play or display the paid answering prompt message. The calling user can reply to the paid answering prompt message. After receiving a payment consent message from the calling terminal indicating agreement to pay the answering fee, the called terminal will trigger the sending of payment timing option information to the calling terminal. After the calling terminal plays or displays the payment timing option information, the calling user can select the desired payment timing option and reply. If the calling user selects the post-pay option, a payment commitment message is generated, and the called terminal triggers a call reminder based on the payment commitment message. Because the payment commitment message indicates the calling user's willingness to pay the answering fee, but the payment has not yet been completed, it is necessary to send a payment instruction message to the calling terminal after the call ends. This not only provides the calling user with payment instructions so that the calling user can pay the answering fee according to the payment instruction message, but also reminds the calling user to pay, reducing the probability that the calling user will forget to pay the answering fee after hanging up the call.

[0174] Furthermore, to prevent the calling user from defaulting on the contract, a subsequent mechanism is required to monitor and verify whether the actual payment is completed on time. If the payment is not completed, appropriate measures need to be taken, such as restricting future call requests or issuing a reminder to the calling user. Therefore, in some embodiments, the payment confirmation information includes committed payment information; accordingly, the method may further include:

[0175] After the call ends, a default handling step is performed based on the committed payment information, wherein the default handling step includes at least one of the following:

[0176] When it is detected that the promised payment information has not been fulfilled within the first time limit, sending a call fee recovery prompt message to the calling terminal, wherein the call fee recovery prompt message is used to prompt the calling user to pay the call fee for this call;

[0177] When it is detected that the promised payment information is not fulfilled within the second time limit, a default information for the calling number is sent to the service end, and the default information is used for other called terminals connected to the service end to restrict the calling number from answering calls.

[0178] The first time limit refers to a shorter, pre-set timeframe within which the calling user must complete payment. Failure to complete payment after the first time limit constitutes a breach of contract. The second time limit refers to a longer, pre-set timeframe. Failure to complete payment after this timeframe constitutes a serious breach of contract. The second time limit is greater than the first time limit. A call fee recovery reminder message is a reminder message sent to the calling user to remind them to pay the call fee. Dishonesty information refers to information recording a calling user's failure to fulfill their payment commitment. It may include a dishonest flag that marks the calling user and / or the calling number as untrustworthy. Users and numbers marked as untrustworthy are referred to as dishonest users and dishonest numbers, respectively. Call restriction refers to restrictions on subsequent calls from dishonest numbers and / or dishonest users. For example, this may include rejecting the current call from the calling number, revoking the calling number's post-payment privileges, requiring the calling user to settle any unpaid call fees before continuing to call, or increasing the call fees required of the calling number. Call restriction on dishonest numbers and dishonest users may be implemented in at least one of the following ways.

[0179] For example, when a calling user dials the called user and selects the postpaid option of "pay after the call", the called terminal will generate a promised payment information and trigger a call reminder to allow the call. After the call ends, it will start monitoring the fulfillment of the promised payment information. If it is detected within the first time limit (for example, 2 hours) that the promised payment information has not been fulfilled, that is, the calling user has not completed the payment, the service end of the called terminal or the first call application will automatically generate a call fee recovery reminder message. The call fee recovery reminder message can be sent to the calling terminal via SMS, email or in-app notification. The content usually includes detailed information of the call, the amount payable, and the payment method and link, etc. The call fee recovery reminder message can clearly inform the calling terminal that the fee needs to be paid as soon as possible to avoid further breach of contract. If it is still detected that the promised payment information has not been fulfilled within the second time limit (for example, 24 hours), a default information can be generated for the calling number. The default information will be stored in the server and shared with other called terminals connected to the server. These called terminals can restrict the answering of calls from the above-mentioned calling number or calling user based on the default information. The restriction measures may include directly rejecting the call, forwarding the call to voicemail, or displaying a warning message when answering the call to remind the called user that this is a default code, etc. It should be noted that the above-mentioned first and second time limits can be flexibly set according to actual needs. Generally, the second time limit only needs to be greater than the first time limit. This application does not limit its specific values.

[0180] Through the above-mentioned method, the payment behavior of the calling user can be effectively managed to ensure that the rights and interests of the called user are protected. By sending the prompt information of the answering fee recovery in time, the calling user can be reminded to complete the payment as soon as possible to ensure the benefit rights of the called user; for users who fail to fulfill their payment commitments for a long time, they can be marked as untrustworthy users, or their calling numbers can be marked as lost numbers, and corresponding restrictive measures can be taken to effectively prevent malicious behavior; a strict breach of contract handling mechanism can enhance the trust of the entire paid answering mechanism, so that more users are willing to use the paid answering function; the entire breach of contract handling process can be implemented through an automated program to reduce manual intervention, improve efficiency and reduce operating costs. Based on the above description, the implementation method of the present application can take effective breach of contract handling measures when the calling user fails to fulfill the payment commitment on time, which can not only protect the rights and interests of the called user, but also maintain the healthy operation of the entire paid answering mechanism, while improving user experience and trust, and realizing an efficient, convenient and safe paid answering process.

[0181] In some modified implementations, before executing the paid answering process, it may also include: detecting whether the calling number is a lost signal code; if it is detected that the calling number is a lost signal code, restricting the calling number from answering the call; if it is detected that the calling number is not a lost signal code, continuing to execute subsequent processes.

[0182] Among them, detecting whether the calling number is a lost signal code can be achieved by sending a lost signal code query request for the calling number to the server. After receiving the number query request, the server can query the database whether the calling number corresponds to lost trust information. If there is lost trust information, the calling number is judged to be a lost signal code. If there is no lost trust information, the calling number is judged not to be a lost signal code, and the judgment result indicating whether the calling number is a lost signal code is fed back to the called terminal, or the query result of whether the calling number corresponds to lost trust information is fed back to the called terminal, and the called terminal judges whether the calling number is a lost signal code based on the query result. The above can all achieve the purpose of the embodiments of the present application and should all be within the scope of protection of the present application.

[0183] It should be noted that the above-mentioned step of detecting whether the calling number is a lost signal code can be set before any step before executing the paid answering process. For example, it can be detected immediately after receiving the call request whether the calling number is a lost signal code, so as to restrict calls with lost signal codes at the beginning of the call. The subsequent process that continues to be executed can be the process after the step of detecting whether the calling number is a lost signal code.

[0184] The above-mentioned call restriction can be implemented by using at least one of the following call restriction methods:

[0185] The first call restriction method is to reject incoming calls. This method allows you to directly reject incoming calls from the calling number, completely preventing the called user from being disturbed by untrusted sources and ensuring that only trusted calls can enter the user's communication environment. It also sends a clear message to the calling user: due to previous unfulfilled payment commitments, the current call attempt has been rejected, which helps encourage the calling user to resolve historical outstanding debts as soon as possible. Furthermore, a call restriction reminder can be sent to the calling terminal through paid call information exchange. This reminder can be used to remind the calling user that they must first pay any outstanding call charges before the call restriction can be lifted.

[0186] The second call restriction method: cancels the post-payment permission for the calling number. This restriction does not support the pay-before-answer model, but instead relies on a pay-before-answer model. The calling user must prepay the call fee as a guarantee to trigger the incoming call reminder on the called terminal. This ensures that the call fee is securely credited before the call begins, reducing the risk of subsequent collection.

[0187] The third call restriction method requires the calling user to settle any outstanding incoming call charges before continuing the call. This method proactively reminds the calling user to pay any outstanding incoming call charges as a prerequisite for continuing the call. This not only aims to recover past debts but also provides the calling user with an opportunity to correct their mistakes, encouraging them to comply with established payment rules. Once the calling user completes payment, their number is removed from the default list, and normal communication rights are restored. Therefore, while protecting the rights of the called user, this method also provides flexibility. It is suitable for situations where the breach of contract may be due to negligence rather than intentional, and helps rebuild trust between the two parties.

[0188] The fourth method of call-answering restriction: increasing the call-answering fee required of the calling number. For callers who frequently breach their contracts or have been repeatedly listed as dishonest, higher charges can be set to increase the cost of repeat calls. This serves as a warning and screens out callers with truly urgent needs who are willing to bear the higher costs. The higher fee threshold effectively reduces unnecessary nuisance calls and protects the privacy and time resources of the called user. At the same time, this arrangement also provides a way for users who occasionally breach their contracts but hope to regain trust: by paying a higher fee, they can prove their sincerity and gradually restore normal communication rights.

[0189] The answering restriction in the embodiment of the present application can be based on the above multiple answering restriction methods, flexibly set in combination with actual needs and expanded on the above basis, and the embodiment of the present application does not limit it. Among them, the above multiple answering restriction methods can be implemented in combination, or they can be implemented in a graded manner according to the degree of restriction, such as the degree of restriction is proportional to the number of breaches of trust or the degree of breach of trust. The more breaches of trust or the higher the degree of breach of trust, the more severe the degree of restriction. Conversely, the fewer breaches of trust or the lower the degree of breach of trust, the lighter the degree of restriction. Among them, the degree of restriction of various answering restriction methods can be flexibly set according to actual needs, and the embodiment of the present application does not limit it. The calling user can remove the breach of trust mark and answering restriction by paying the historical unpaid answering fees, or reduce the degree of answering restriction by increasing the number of fulfillment times or the fulfillment ratio, etc. All of the above can urge the calling user to pay the answering fee in time and ensure that the called user can receive the answering compensation.

[0190] It should be noted that the above-mentioned payment confirmation information only needs to have a corresponding relationship with the calling number. The embodiment of the present application does not limit the payment entity for the answering fee. The payment entity can be the calling terminal, the payment device or the server, etc. Accordingly, the above-mentioned payment confirmation information can be generated based on the payment behavior of the calling terminal, or it can be generated based on the payment behavior of other payment devices, or it can be generated by the server based on the deduction agreement corresponding to the calling number from the designated account, or it can be generated based on the automatic payment mechanism payment, etc. The embodiment of the present application does not make any limitation, as long as the payment confirmation information is associated with the calling number and can indicate that the payment is made for the call behavior of the calling number. In addition, the payment confirmation information and the calling number can be directly or indirectly associated. For example, when the calling user makes a payment, the second call application can automatically add the calling number as payment note information. After the payment is completed, the payment confirmation information carries the above-mentioned payment note information. The called terminal can determine that the payment completion information is associated with the calling number based on the payment note information; for example, the called terminal or server can generate a payment order for the calling number based on the call request. After the calling terminal makes payment according to the payment order, the payment confirmation information can be indirectly associated with the calling number based on the payment order; the above is only an exemplary explanation of associating the payment confirmation information with the calling number. Those skilled in the art can implement it directly or after modification, and it should all be within the scope of protection of this application.

[0191] In addition, in cases where the called terminal supports multiple card services, in order to make the payment purpose of the payment confirmation information clearer, the payment confirmation information can also be associated with the called number, for example, by adding the called number in the above-mentioned payment note information to directly associate the payment confirmation information with the called number, or indirectly associate the payment confirmation information with the called number through a payment order, etc., all of which should be within the scope of protection of this application.

[0192] It should be noted that the calling number can be a mobile phone number, a fixed number, a telemarketing number, a virtual number or a special number. The embodiments of this application do not limit this, and they are all applicable to the call processing solution provided in this application.

[0193] In some application scenarios, the calling number may be a fixed number, and accordingly, the calling terminal may be a communication device such as a fixed telephone that does not have independent payment capabilities. In order to realize the payment of the answering fee in this case, in some embodiments of the present application, a payment agent is supported to pay the answering fee for the calling terminal that does not have independent payment capabilities. Accordingly, in the pay-first-then-answer mode, the payment confirmation information may be a payment completion information generated based on the answering fee paid by the payment agent, so that the call to the called number can still be realized even if the calling terminal does not support the payment function; In addition, the payment agent can also pay for the calling terminal after the call ends in the answer-first-then-pay mode, which is also within the scope of protection of this application. It should be noted that the payment agent provided in the embodiments of the present application can be a terminal device such as a mobile phone, tablet computer, laptop computer, desktop computer, etc., or a server device such as a server, etc. It can be a device with payment function set up by the calling user, such as a call center management machine, a backend service device of call software, etc., or it can be a payment agent provided by the server side specifically for providing payment services to users, etc., and the embodiments of the present application are not limited thereto.

[0194] In order to simplify the payment process and improve the user experience, the embodiment of the present application also provides an automatic payment mechanism. After determining that the incoming call is a harassing call, the corresponding answering fee can be deducted from the payment account pre-bound to the calling number directly or after confirmation by the calling user. The calling user does not need to pay manually each time. After the automatic payment is completed, the payment confirmation information can be triggered to further trigger the call reminder of the called terminal.

[0195] In one example, the administrator of a calling number can enable the automatic payment feature for the calling number and bind a payment account based on an automatic payment agreement. This payment account can be a bank account, credit card account, third-party payment platform account, or a calling platform account opened on the calling application server, though this is not limited in this embodiment of the present application. Individual users can enable the automatic payment feature and bind a payment account in a second calling application. For corporate users, an administrator can enable the automatic payment feature and bind a payment account by signing an automatic payment agreement through a management webpage or client provided by the server. The same payment account can be bound to, but not limited to, one phone number. Multiple phone numbers can be bound to the same payment account to meet company requirements for unified payment of incoming calls initiated by employees. For example, for a company-based telesales platform or team with centralized billing, multiple phone numbers of the telesales team can be bound to the company's centrally managed payment account. Furthermore, the same phone number can be bound to, but not limited to, one payment account, including, for example, a primary payment account and a backup payment account, to ensure smooth payment of incoming call charges. After the automatic payment feature is enabled, the binding relationship between the calling number and the payment account will be stored by the server. After the called terminal receives the call request, if it determines that the incoming call is a nuisance call, it can trigger an automatic payment step. Accordingly, the paid answering process can also include: an automatic payment step; the automatic payment step includes: sending an automatic payment request to the server, the automatic payment request is used to request the server to automatically pay the answering fee from the payment account bound to the calling number according to the automatic payment agreement corresponding to the calling number. After the server receives the automatic payment request, it can automatically pay the answering fee from the payment account bound to the calling number according to the automatic payment agreement corresponding to the calling number. It should be noted that the automatic payment request can be sent by the called terminal, it can also be sent by the calling terminal, and it can also be automatically triggered by the server, all of which can achieve the purpose of the embodiments of the present application.

[0196] The automatic payment mechanism is applicable to various call scenarios, especially those that require frequent paid answering and have a stable source of payment. The automatic payment mechanism can significantly improve the efficiency and convenience of the paid answering process, provide a smoother and undisturbed call experience, and thus optimize the user experience. It can not only reduce the trouble of the calling user having to perform additional payment operations before each call, but also ensure that the payment process is fast and accurate, and avoid the risk of call interruption due to delayed payment. In addition, for common telecommunications call scenarios such as virtual numbers, landlines, company phones, and telemarketing platforms, the payment of answering fees is affected by the fact that the calling terminal does not have a payment function, the calling user is not convenient to pay, and the company or unit needs to pay in a centralized manner or pay on behalf of the calling user. These can be effectively solved through the automatic payment mechanism or the payment on behalf of the calling user, so that the answering fee can be paid smoothly, ensuring that the present application method can be applied to various call scenarios.

[0197] It should be noted that in order to avoid the abuse of the automatic payment mechanism, in some embodiments, it is necessary to trigger the automatic payment operation after the calling user confirms. Accordingly, the above-mentioned automatic payment steps may specifically include: detecting whether the calling number is bound to a payment account for automatic payment; when it is detected that the calling number is bound to the payment account, sending automatic payment option information to the calling terminal, and the automatic payment option information is used for the calling user to choose to use the automatic payment mechanism to pay the answering fee; after receiving the automatic payment consent indication information returned by the calling terminal indicating that the calling user agrees to pay the answering fee through automatic payment, sending an automatic payment request to the server.

[0198] Among them, detecting whether the calling number is bound to a payment account for automatic payment can be achieved by sending a query request to the server. Since the server stores a large number of binding relationships between phone numbers and payment accounts, by querying the above binding relationships, it is possible to detect whether the calling number is bound to a payment account. If no payment account is bound, paid answering information interaction can be triggered to continue the paid answering process; if a payment account is bound, automatic payment option information can be triggered to be sent. After the calling terminal receives the automatic payment option information, the automatic payment option information will be displayed or played in voice for the calling user to choose. The calling user can choose whether to make automatic payment. If automatic payment is selected, an automatic payment request can be triggered to be sent to the server. If automatic payment is not selected, paid answering information interaction can be triggered to continue the paid answering process.

[0199] Considering that the calling user necessarily expects to obtain sufficient talk time when agreeing to pay the answering fee, if the called user hangs up immediately after the call is connected, this behavior obviously does not meet the expectations of the calling user. Frequent occurrence will gradually erode the enthusiasm of some calling users to pay, which is not conducive to establishing a fair, efficient and sustainable communication environment. Therefore, it is necessary to reasonably restrict the answering time of the called user to ensure that the rights and interests of the calling user are fully protected, thereby balancing the rights and obligations of the calling user and the called user in the paid answering scenario and maintaining a good call experience. Therefore, based on any of the aforementioned embodiments, in some modified embodiments, the method may also include:

[0200] After detecting that the called party answers the call, the answering time protection process is executed. The answering time protection process includes using a preset answering time protection method to protect the answering time of the call. The answering time protection method includes at least one of the following:

[0201] Play or display a call protection prompt message, wherein the call protection prompt message includes a prompt message prompting the called user that the answering time must be greater than the call protection time;

[0202] Displaying the call duration progress information on the call interface of the called terminal, the call duration progress information including the call timing information and the call protection duration indicator;

[0203] During the call protection period, in response to the first hang-up trigger input by the called user, the call status is maintained and the call protection prompt information is displayed;

[0204] During the call protection period, in response to the second hang-up trigger input by the called user, the call is hung up;

[0205] The hang-up control is hidden within the answering protection time, and is displayed after the answering time exceeds the answering protection time. The hang-up control is used for the called user to input a hang-up trigger operation.

[0206] Among them, the answering time protection process refers to a series of measures that are initiated according to preset conditions after detecting that the called user answers the call, in order to prevent the called user from ending the call prematurely and ensure that the call can meet the preset minimum time requirement. These measures may include but are not limited to playing or displaying answering protection prompt information, displaying answering time progress information, responding to the first hang-up trigger operation to maintain the call status and prompt, hiding the hang-up control, etc., aiming to protect the rights and interests of the calling user after payment, and avoid unnecessary fee disputes or service quality issues due to short calls. Taking into account that the answering time protection process mainly protects the rights and interests of paid calling users, the answering time protection process can be triggered and executed when the payment confirmation information is payment completion information. When the payment confirmation information is committed payment information, the answering time protection process may not be executed, or the answering time protection process may be executed. The specific settings can be made by the user and are not limited in the embodiments of this application.

[0207] The answering protection time is a minimum call duration standard set by the user or system by default in the paid answering service, such as 30 seconds, 1 minute, etc., and is not limited in the embodiments of the present application. The measurement time starts from the time the called user answers the call, and the artificial intelligence voice interaction stage is not included. It aims to ensure that the calling user can get enough time to communicate with the called user, and avoid poor user experience or damage to rights due to too short call time. After the calling user pays the answering fee, the called terminal or service end can judge whether the actual call has met the reasonable duration requirement based on the answering protection time. If the called user answers the call but the actual answering time does not reach the answering protection time, the refund processing process can be executed according to the set rules to refund part or all of the fees to the calling user to compensate for the failure to obtain the expected service experience.

[0208] The answering time progress information is the information displayed on the call interface of the called terminal, which includes the answering timing information and the answering protection time mark of the current call, and is used to remind the called user of the current call time and the minimum call time required. The answering protection time mark can mark the answering protection time through visual information (such as images, text, icons, color bars, progress bars, etc.), so that the called user can intuitively understand the shortest call time required. Combined with the answering timing information, the answering time progress information can be displayed, thereby intuitively conveying the time progress information of the current call to the called user, thereby preventing the called user from hanging up the phone prematurely to a certain extent, and protecting the rights and interests of the calling user.

[0209] The answering protection prompt information is a prompt information used to remind the called user that the answering time must be greater than the answering protection time. It can be played through voice or displayed on the screen. It can be played or displayed just after the called user answers the call, or it can be played or displayed after the called user enters the first hang-up trigger operation, so as to provide timely reminders when the called user tries to hang up.

[0210] The hang-up control is an interface element for the called user to input the hang-up trigger operation. It can be displayed continuously during the call, or it can be hidden during the answering protection time according to the user's settings. It will be redisplayed after the answering time exceeds the protection time to allow normal hang-up operation. By hiding the hang-up control, the difficulty and complexity of the called user hanging up the call can be increased, thereby more effectively preventing the called user from hanging up the call prematurely.

[0211] In some examples, after the calling user pays the call fee, the called terminal will receive an incoming call notification. After the called user answers the call, the called terminal's call interface will display a call protection prompt to remind the called user that they need to answer the call for a sufficient length of time to receive the call fee. The call interface will also display call progress information, including the call timer and the call protection time indicator. This approach not only allows the called user to clearly understand the timeline of the current call but also reminds them of the minimum call duration required. If the called user attempts to end the call by the first hang-up trigger within the call protection time, the called terminal can maintain the call and display a call protection prompt on the screen, clearly informing the called user that they must continue the call until the protection time expires. If the called user attempts to hang up again (i.e., the second hang-up trigger), the call can be allowed to terminate, protecting the user's right to freedom of communication and initiating an automatic refund process. Although this may reduce the called user's call revenue, it satisfies the called user's right to hang up at any time and avoids the inconvenience or dissatisfaction caused by forced call answering. In addition, to further prevent accidental operations or intentional hang-ups, the hang-up control can be hidden within the call protection time. The hang-up control will only be redisplayed after the call duration exceeds the protection time, allowing the called user to end the call normally, thereby further ensuring the rights and interests of the calling user.

[0212] It should be noted that the answering time, call answering timing information, etc. in the embodiments of the present application are all counted from the time when the called user is detected to answer the call, rather than from the time when the artificial intelligence voice assistant answers the call, so as to accurately record the actual call duration.

[0213] This implementation, by introducing a call duration protection mechanism, ensures that every paid call meets the expectations of the calling user, preventing dissatisfaction among the called user due to premature hang-ups. Therefore, this approach not only protects the rights and interests of the calling user and strengthens their trust in paid call answering services, but also effectively balances the interests and needs of both the calling and called user. While protecting the call duration, it also fully supports the called user's ability to hang up at any time, ensuring that every paid call meets the reasonable expectations of both parties, maintaining a fair and transparent paid call environment, and improving the user experience.

[0214] In some modified implementations, after detecting that the called user answers the call, executing the answering duration protection process may include:

[0215] If the calling number meets the answering time protection condition, the answering time protection process is executed after detecting that the called user answers the call; wherein the answering time protection condition includes at least one of the following:

[0216] The payment confirmation information corresponding to the calling number is the payment completion information;

[0217] The answering fee paid by the calling number includes the answering time protection fee;

[0218] The calling number belongs to the preset number category that supports call duration protection;

[0219] The membership level corresponding to the calling number belongs to the preset membership level that supports call duration protection;

[0220] The calling number has enabled call duration protection.

[0221] Among them, the answering time protection condition is the specific condition for determining whether the answering time protection process needs to be initiated, which can be determined based on multiple dimensions such as payment status, payment content, number category, membership level and permission settings.

[0222] The answering time protection fee refers to an additional fee paid by the calling user when paying the answering fee to ensure that the call can meet the preset minimum duration requirement. It is usually included as part of the answering fee. For example, the answering fee can include a basic fee and an additional fee. The basic fee is the necessary fee for triggering the call reminder, while the additional fee is the fee for obtaining additional benefits. Additional benefits include answering time protection, personalized reminder ringtones, personalized reminder effects, etc. The fee for obtaining the answering time protection benefit can be called the answering time protection fee. When the calling user chooses to pay the answering fee that includes the answering time protection fee, the called terminal can automatically initiate the answering time protection process after the called user answers the call if it detects that the answering fee includes the answering time protection fee. If it detects that the answering fee paid by the calling user does not include the answering time protection fee, the answering time protection process may not be executed. By explicitly confirming payment completion and triggering the call protection process only when the payment includes a call protection fee, the call protection process is activated only if the caller agrees and has paid the fee. This ensures that the call protection process is initiated based on the caller's needs, giving the caller full autonomy and meeting their individual needs. Furthermore, the call protection fee provides additional benefits to the called user, ensuring a balanced return. If the called user needs to terminate the call early, the fee can be refunded through a subsequent refund process, ensuring the rights and interests of both the caller and the called user are protected. By dividing the call fee into a base fee and an additional fee, the system not only provides flexibility to meet diverse user needs but also improves overall service quality and user experience. For standard calls, callers only need to pay the base fee for basic paid call service. Callers with specialized needs can choose to pay additional fees for additional value-added services and safeguards. This approach not only enhances flexibility and adaptability, but also provides users with more choices, so that each call can obtain the optimal service configuration based on actual conditions.

[0223] Number categories that support answering time protection refer to preset number categories that meet specific conditions. These number categories can support the initiation of the answering time protection process after answering an incoming call. Different categories of numbers (such as unclassified calls, marketing calls, fraudulent calls, etc.) may have different additional fee charging standards and service terms, so whether to start the process can be decided based on the number category. For example, unclassified calls can support answering time protection, so the answering time protection process can be executed; while fraudulent calls should not be protected by answering time and do not support answering time protection, so the answering time protection process should not be executed. By checking the category of the calling number, the answering time protection mechanism can be flexibly applied to adapt to different usage scenarios and technical requirements. This method can not only improve the effectiveness of the call, but also protect the rights and interests of the calling user.

[0224] Membership levels that support call duration protection automatically determine whether to initiate call duration protection based on the calling user's membership level. Premium members or users who choose specific service packages may enjoy additional protections and service guarantees, thereby improving user satisfaction. For example, premium members are entitled to call duration protection to ensure the quality and effectiveness of their important calls. By checking the membership level corresponding to the calling number, a personalized service experience can be provided based on the user's subscription level. This approach not only enhances the user experience but also enables differentiated services based on the actual needs of different users, improving overall service satisfaction.

[0225] Call duration protection permissions determine whether the calling number has enabled call duration protection. Only if this permission is enabled will the corresponding call duration protection process be initiated to protect the rights of the calling user. This permission setting can be enabled or disabled by the user, or configured by the service provider based on the user's subscription level or specific needs. By checking whether the calling number has enabled call duration protection permissions, the call duration protection process is ensured only if the calling user has explicitly agreed and enabled this feature.

[0226] During specific implementation, it is possible to first detect whether the calling number meets any of the answering time protection conditions. If any of the conditions are met, the answering time protection process is immediately started. For example, when the payment confirmation information of the calling number is payment completion information, and the answering time protection fee is included in the paid answering fee, the protection process can be automatically started after the call is answered. Similarly, if the number category of the calling number belongs to the preset number category that supports answering time protection, or the membership level corresponding to the calling number belongs to the preset membership level that supports answering time protection, the answering time protection process can also be started. In addition, it is also possible to check whether the calling number has activated the answering time protection permission to ensure that the protection process is only started when the permission has been activated.

[0227] It should be noted that the above-mentioned multiple answering time protection conditions can be used one by one, or multiple conditions can be selected and used in combination. When used in combination, it can be set to execute the answering time protection process when any one item is met, or it can be set to execute the answering time protection process when all items are met. The embodiment of this application does not limit this, and technical personnel in this field can flexibly implement it according to actual needs, which should all be within the scope of protection of this application.

[0228] By introducing the above-mentioned answering time protection conditions, it is possible to determine whether to execute the answering time protection process based on the answering time protection conditions, thereby safeguarding the rights and obligations of both the calling user and the called user. Moreover, the answering time protection mechanism can be flexibly applied according to conditions in different dimensions, adapting to various complex scenarios and technical requirements, meeting users' diverse and personalized paid answering needs, enhancing user experience, and enabling users to pay more attention to effective communication during calls, thereby improving overall service satisfaction.

[0229] Considering that, for situations where the calling number is a friend's new number or a new customer's number, the AI ​​voice interaction stage may make a misjudgment, resulting in the other party paying the answering fee before answering the call. In addition, there are also situations where the calling user has paid but the called user does not answer the call. In these situations, it is often necessary to refund the paid answering fee to the calling user. In view of the above requirements, in some modified implementations, the method may further include:

[0230] When the refund trigger conditions are met, the refund processing flow is triggered, and the refund processing flow includes a full or partial refund of the answering fee paid by the calling user; wherein, the refund trigger conditions include at least one of the following: the call is not answered by the called user, the call is ended by the called user and the answering time is less than the preset answering protection time, and a manual refund operation input by the called user is detected.

[0231] Among them, the refund triggering conditions refer to the specific conditions that can start the refund processing process, ensuring that part or all of the answering fees can be refunded automatically or according to user operations under specific circumstances. The answering protection time is a pre-set minimum call duration standard to protect the rights and interests of calling users and prevent too short call time from affecting the user experience. The manual refund operation refers to the process in which the user actively initiates and requests the execution of the refund processing process through a specific interface or channel. The refund processing process refers to a series of processes that are automatically executed by the called terminal or the service end or according to the user's request on the basis of the paid answering service when specific refund triggering conditions are met. The purpose is to refund part or all of the answering fees paid by the calling user, or to cancel the payment commitment information made by the calling user and no longer charge the calling user the answering fees that he or she has committed to pay.

[0232] The refund process involves multiple steps and technical mechanisms to ensure a secure, accurate, and efficient refund process. When a refund trigger condition is detected, the refund process is initiated. For example, the refund request is first verified to confirm whether it complies with the pre-set refund rules. The refund amount is then calculated. This amount may be a full refund or a proportional partial refund, depending on the actual call situation and the pre-set refund rules. The refund is then returned to the caller's payment account through a secure payment channel to ensure the security and accuracy of the funds. Relevant information about the refund event, including the reason, amount, and time of refund, is recorded for subsequent audit and analysis to improve overall service quality. Furthermore, to enhance user experience and system reliability, the refund process may also include notifications to users regarding the refund status and results. This not only enhances users' right to be informed but also strengthens their trust in paid answering services. For example, when a refund is successfully completed, an SMS or in-app message can be sent to the caller, stating the refund amount and estimated arrival time, ensuring that users are promptly informed and confirmed of the refund.

[0233] In some examples, when the calling user initiates a call but the called user does not answer the call and automatically hangs up after a certain period of time, the called terminal or the server can automatically trigger a refund process to refund the paid answering fee to the calling user in full or in proportion to protect the rights and interests of the calling user. If the call is connected but the actual call duration does not reach the preset answering protection time (for example, 1 minute), the refund process can also be executed according to the set rules to ensure that even if the call is short, the calling user will not suffer unnecessary economic losses due to the short call time. It should be noted that this method can limit the refund to only the case where the called user triggers the hang-up call, and no refund is required for the case where the calling user hangs up the call. In addition, in order to give the called user a certain degree of flexibility and control, it is also allowed to start the refund process by manually triggering the refund control to meet the called user's need to actively refund the answering fee.

[0234] This implementation method can support refund demands in various scenarios by introducing multiple refund trigger conditions. It can not only protect the rights and interests of calling users, but also meet the needs of called users for independent refunds, ensuring that every paid call meets the reasonable expectations of both parties as much as possible.

[0235] Among them, the refund amount of the above refund processing flow can be a full refund, or it can be a partial refund according to the platform regulations or the customized requirements of the called user. The embodiments of this application are not limited to this, and they should all be within the scope of protection of this application.

[0236] The above refund processing flow can be initiated by the called terminal. For example, for the answering fee paid directly to the called user's account, the called terminal can initiate a refund; in addition, the above refund processing flow can also be initiated by the server. For example, for the implementation method in which the answering fee is first paid to the server and then paid to the called user by the server after the call ends, the server can initiate a refund after detecting that the refund trigger conditions are met after the call ends, and no longer pay the called user; in addition, it can also be that the called terminal requests the server to trigger, etc., all of which can achieve the purpose of the embodiments of the present application, and will not be described one by one here, and they should all be within the scope of protection of the present application.

[0237] In some examples, the calling user pays the answering fee to the service end (such as a payment platform), and the service end generates a payment confirmation message and sends it to the called terminal. The called terminal triggers a call reminder based on the payment confirmation message. After the call is completed, if the answering time is longer than the answering protection time, the remaining answering fee will be paid to the called user's first account after deducting part of the platform technical service fee from the answering fee. If the answering time is less than the answering protection time, the refund processing process will be executed to refund the answering fee in full or in part to the calling user.

[0238] Based on the above embodiment, in some modified embodiments, the method may further include:

[0239] After the call ends, a call end interface is displayed, and a manual refund trigger element is displayed on the call end interface, wherein the manual refund trigger element is used for the called user to input a manual refund operation;

[0240] In response to a manual refund operation input by the called user based on the manual refund trigger element, it is determined that a refund trigger condition is met.

[0241] The call end interface refers to the screen or page that the user sees after a call ends. This interface not only summarizes the call information that just ended, such as the call duration and cost details, but also provides a variety of follow-up operation options, such as refund options, adding contacts options, and tagging options. These operation options can be displayed and selected by the user using elements such as controls.

[0242] On the call end interface, the manual refund trigger element is a specific user interface element used to provide the called user with a refund option and for the user to enter a manual refund operation. It may specifically include buttons, selection forms, controls, input boxes or other interactive components, and may be attached with a refund mark (for example, "one-click refund of answering fee", "drag the answering fee mark for refund", etc.) for user identification. When the called user clicks, selects or drags the manual refund trigger element, it means entering a manual refund operation, and the called terminal can trigger the refund processing flow. In the refund processing flow, the refund can be directly performed according to predetermined rules. In this case, the overall refund process can be called a one-click refund, that is, the called user only needs to click, select or drag the refund option in the manual refund trigger element, and the refund can be automatically completed without subsequent operations.

[0243] In some implementation variations, the refund option displayed on the call end interface is not limited to one, but may be multiple. Different refund options correspond to different refund amounts. The refund amount may be displayed as a refund ratio or as a specific amount. This is not limited in the present embodiment, such as 100% refund, 80% refund, 50% refund, 30% refund, 1 yuan refund, 0.5 yuan refund, etc. Multiple refund options may be displayed and accepted by the user using multiple manual refund trigger elements as carriers, that is, the refund options and the manual refund trigger elements are displayed one by one. Corresponding; In addition, multiple refund options can also be integrated into a manual refund trigger element for display and user selection. For example, the manual refund trigger element can use a drop-down menu, check box, selection form, etc. to display multiple refund options for the called user to select; In addition, the manual refund trigger element can also include a refund amount input box, and the called user enters the refund amount in the refund amount input box; In addition, the above refund options and refund amount input box can also be separated from the manual refund trigger element and set on the call end interface, which can achieve the purpose of the embodiment of the present application. In summary, the above refund processing flow can specifically include refunding the call fee paid by the calling user according to the refund amount input (including selection) or confirmed by the called user, thereby providing users with a convenient and fast manual refund solution that supports customized refund amounts, meeting users' diverse and free manual refund needs. Among them, the above-mentioned refund amount can be input based on the manual refund trigger element, for example, the above-mentioned refund options correspond one-to-one with the manual refund trigger element and multiple refund options are integrated into one manual refund trigger element, thereby combining the input of the refund amount with the manual refund trigger element. On the one hand, it simplifies the display interface, and on the other hand, it is more convenient for users to quickly enter the required refund amount for manual refund operations, which helps to improve the efficiency of manual refunds and user operation experience.

[0244] Based on the above embodiment, in some modified embodiments, the method may further include:

[0245] After the call ends, check whether the call duration has reached the preset call protection time;

[0246] When the duration of the current call is less than the call protection time, it is determined that the refund triggering conditions are met.

[0247] After a call ends, the actual call duration can be automatically detected and compared with the preset call protection time. If the detection result shows that the call duration is less than the preset call protection time, it is automatically determined that this situation meets the refund trigger conditions and the corresponding refund processing process is initiated. Through this embodiment, it can ensure that the refund processing process is automatically initiated when the call duration is shorter than the preset call protection time, so that the rights and interests of the calling user can be effectively protected.

[0248] Based on the above embodiment, in some modified embodiments, the method further includes:

[0249] When it is detected that the incoming call is not answered by the called user, it is determined that the refund triggering condition is met.

[0250] Among them, not being answered by the called user may mean that the called user did not answer the call or directly refused to answer the call within a period of time after the call reminder. If the calling user has paid the answering fee, this obviously meets the refund triggering conditions and a refund can be made.

[0251] Through this implementation, it is possible to automatically detect and respond to situations where calls are not answered without the need for additional user operations, thereby simplifying the refund application process and ensuring that the refund processing process is automatically initiated when the call is not answered, so that the rights and interests of the calling user can be effectively protected.

[0252] It should be noted that the refund mentioned in the above-mentioned embodiment description can be implemented when the payment confirmation information is payment completion information, thereby refunding the answering fee paid by the calling user. In addition, it can also be implemented when the payment confirmation information is committed payment information. In this case, the refund can be understood as the return of the calling user's payment commitment, and the calling user will no longer be charged the answering fee that he has promised to pay.

[0253] It should be noted that during the artificial intelligence voice interaction process, the called terminal can display artificial intelligence voice interaction prompt information to inform the called user that it is currently conducting artificial intelligence voice interaction with the calling user, so that the called user can understand the incoming call information and its processing status in real time. However, in actual applications, the artificial intelligence voice interaction prompt information may cause information harassment to the called user. Most called users prefer to let the called terminal automatically complete the artificial intelligence voice interaction process in the background without real-time intervention in the process. Therefore, based on any of the aforementioned implementations, in some modified implementations, answering the incoming call and conducting artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant may include: answering the incoming call in the background and conducting artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant, and answering the incoming call in the background includes answering the incoming call while disconnecting the external voice collection link and call playback link of the called terminal.

[0254] Among them, the background involved in the embodiment of the present application is a concept relative to the foreground. The foreground is the part that the user can directly see and interact with. It is mainly responsible for displaying the interface to the user, receiving user input and other related tasks. The user has a direct perception of the foreground tasks. The background is the part that the user cannot see and does not require user interaction. It is mainly responsible for data processing, logical reasoning, automated execution and other related tasks. The user is generally unaware of the background tasks. The background call connection in the embodiment of the present application refers to establishing a communication connection with the calling terminal in the background of the called terminal. After the connection is made, there is no need for the called user to actually answer the call and talk. Since the action of connecting the call is performed in the background of the terminal, it is called background call connection. After the call is connected in the background, the called terminal will not display or play the call information, and the called user will not be aware of the artificial intelligence voice interaction process.

[0255] It should be noted that if the user has the right to answer incoming calls, background call answering can include the called terminal automatically answering the call without the called user having to manually answer the call. If the user does not have the right to answer incoming calls, background call answering can also include the called terminal answering the call upon detecting an AI-powered answering operation input by the called user.

[0256] When an incoming call is connected in the background, the foreground does not need to remind the user that the call is connected, nor does the called user need to talk to the calling user, and the voice data sent by the calling user through the call link does not need to be played. Therefore, in some embodiments, connecting the incoming call in the background may include connecting the incoming call while disconnecting the external voice collection link and call playback link of the called terminal.

[0257] Among them, the external voice acquisition link refers to the entire path from capturing external sound signals from the microphone and transmitting them to the processing unit (such as the processor of a terminal such as a mobile phone). It is responsible for collecting and sending the voice data of the called party. This path includes hardware and software components to ensure that the captured external sound signals can be processed and used by the terminal. In the present application, disconnecting the external voice collection link of the called terminal may refer to disconnecting any node of the external voice collection link of the called terminal, such as turning off the microphone, or prohibiting the transmission or processing of the audio signal collected by the microphone, etc. Therefore, disconnecting the external voice collection link may include disconnecting the microphone voice collection link. The above-mentioned microphone may include a built-in microphone configured by the called terminal itself, or may include an external microphone connected to the called terminal by wire or wireless means, so that after the incoming call is connected in the background, the voice of the called user side will not be collected or the voice of the called user side will not be sent to the calling terminal, thereby avoiding the sound of the called user side from being leaked to the calling user without the knowledge of the called user, thereby realizing the privacy and security protection of the called user; at the same time, paid answering information is sent to the calling terminal corresponding to the calling number through the call link. In other words, the embodiment of the present application uses the call link to send paid answering information to the calling terminal, thereby achieving the purpose of prompting the calling user to pay the answering fee.

[0258] Specifically, disconnecting the external voice collection link can be achieved at different levels through a variety of technical means, ensuring that the called terminal's microphone does not capture and upload any sound input. For example, at the software level, the First Call application can be programmed to automatically disable the microphone recording function when an incoming call is connected in the background, or temporarily change the audio capture status through the API interface provided by the operating system, such as muting the microphone or stopping the audio service. The microphone's operating mode can also be adjusted through the audio management function to prevent it from receiving sound input, or even redirect the input to an invalid device by modifying the audio routing policy. Through these measures, not only can the collection and transmission of voice data be effectively controlled without affecting other functions, but the privacy of the called user can also be fully protected during the process of connecting the incoming call in the background and sending the paid answering information to the calling terminal, thereby ensuring the security and reliability of communication and enhancing the called user's trust in the paid answering service.

[0259] The call playback link refers to the data transmission path from the communications network to the called terminal's speaker, responsible for receiving and playing the caller's voice data. Disabling the call playback link can be achieved through various technical means at different levels to ensure that the audio from the calling terminal is not played back through the called terminal's speaker. For example, at the software level, the primary call application can be programmed to automatically suspend audio output services when an incoming call is connected in the background. Alternatively, the operating system's API can be used to temporarily change the audio output path, such as redirecting the audio to a muted device or completely disabling audio output. At the hardware level, a dedicated hardware switch or circuit can be built in to directly disconnect the audio signal line to the speaker. Alternatively, the input signal can be ignored by adjusting the operating mode of audio codecs that support dynamic configuration. At the network protocol level, SIP signaling can be used to send appropriate instructions to temporarily suspend one party's audio stream without interrupting the entire call connection. Whether using software control, hardware operation, or network protocol intervention, the core goal is to effectively prevent the audio from the calling terminal from being played back through the called terminal without affecting other functions, thereby enabling the backstage connection of incoming calls and the sending of paid answer information. The above-mentioned speaker may include a built-in speaker configured in the called terminal itself, or may include an external speaker connected to the called terminal via a wired or wireless manner.

[0260] It should be noted that, in this embodiment, the incoming call is connected in the background. Since the incoming call has been connected in the background and the external voice collection link and the call playback link are in a disconnected state, after the incoming call reminder is triggered subsequently, after the called user's answering operation is detected, it is also necessary to restore the external voice collection link and the call playback link to a smooth state, for example, turn on the microphone and speaker, etc., to ensure that the called user can smoothly talk to the calling user after performing the answering operation.

[0261] In some modified implementations of the embodiments of the present application, the method may further include:

[0262] The call reminder interface and / or the call pre-reminder interface displays additional call information and / or call intention about the calling number, wherein the additional call information includes the identity information of the calling user and / or call description information for this call.

[0263] The above-mentioned additional call information can be sent by the calling terminal to the called terminal, or by the server at the request of the called terminal, both of which can achieve the purpose of the embodiments of the present application. The called terminal can directly display the above-mentioned additional call information, or it can first identify the calling user's calling intention based on the above-mentioned additional call information and then display the identified calling intention. When displaying, it can be displayed in the form of a call information card in a fixed format. Formatting the card information helps to improve reading efficiency, thereby helping the called user to quickly understand the identity of the caller or the purpose of the call when viewing the call reminder, and then decide on the answering strategy.

[0264] Among them, the identity information of the calling user can be transmitted to the called terminal through an electronic business card, that is, the above-mentioned additional information of the incoming call may include an electronic business card that records the identity information of the calling user. The electronic business card is a digital business card format, which usually contains detailed information of the user, such as name, telephone number, email address, company name and other identity information. The call description information includes the business information related to the current call entered by the calling user (such as the purpose of the call, appointment matters, etc.). The incoming call reminder interface is a prompt interface displayed on the called terminal when the incoming call reminder is triggered, which is used to notify the user of a new incoming call. The incoming call pre-reminder interface is an interface displayed to the user in advance before the formal incoming call reminder. It can be a prompt interface displayed on the called terminal when the incoming call pre-reminder is triggered, which is used to provide the user with an earlier incoming call reminder.

[0265] For example, when receiving an incoming call request, the called terminal may also receive additional call information sent by the calling terminal or the server. The additional call information may include the identity information of the calling user (such as name, position, company) and / or the service information of this call (such as the purpose of the call, the appointment time), and then display this information on the incoming call reminder interface and / or the incoming call pre-reminder interface. For example, on the incoming call reminder interface, in addition to displaying the caller number, it can also display the name, company name, and avatar of the calling user, as well as a brief description of the call service. If there is a pre-reminder interface for incoming calls, this information can also be displayed in advance on the interface so that the user has more time to prepare for the call.

[0266] The above-mentioned additional call information can be actively sent by the calling terminal. For example, when the calling terminal initiates a call, the calling terminal can directly send the pre-stored electronic business card to the called terminal, so that the called user can quickly understand the background and purpose of the call and make a decision to answer the call. For example, in order to reduce unnecessary data transmission, it can also be determined first whether it is the first time to call the called number, and when it is the first time to call the called number, the pre-stored electronic business card is sent to the called terminal. In other cases, it is sent on demand, that is, it is sent at the request of the called terminal. Since the called user is unlikely to recognize the calling number in the case of the first call, it can help the called user to obtain information about the caller. In other cases, it is sent on demand and not sent without request, which can effectively reduce unnecessary electronic business card sending. For example, when the calling parties have talked many times, Obviously, there is no need to send an electronic business card when initiating a call again. Therefore, this implementation method can not only provide the called user with necessary information support for answering the call, but also reduce unnecessary information sending processes, save network resources and system overhead, and reduce the interference caused to the called user by sending unnecessary electronic business cards. In addition, the called terminal can be queried to see whether the calling number meets the paid answering conditions corresponding to the called number. If so, an electronic business card is sent to the called terminal. This method can more accurately know whether the called user needs the calling user's electronic business card, and send the electronic business card if necessary, otherwise it does not need to be sent. Similarly, it can not only provide the called user with necessary information support for answering the call, but also reduce unnecessary information sending processes, save network resources and system overhead, and reduce the interference caused to the called user by sending unnecessary electronic business cards. In addition, if the calling user uploads the electronic business card to the server in advance, then the above-mentioned methods of sending the electronic business card by the calling terminal can also be changed to the calling terminal triggering the server to send the electronic business card to the called terminal, or the server can identify various scenarios or send the electronic business card to the called terminal according to the setting rules of the calling terminal. In addition, the called terminal can also initiate an electronic business card acquisition request to the server or the calling terminal, and then the server or the calling terminal responds to the electronic business card acquisition request and sends the electronic business card to the called terminal. Those skilled in the art can refer to the above-mentioned exemplary descriptions for flexibly changing the implementation, and will not go into details here.

[0267] The above call intent can be identified based on additional call information or analyzed through AI voice interaction content, which will not be discussed here. By displaying the call intent, it can also help callers quickly understand the background and purpose of the call, improving the accuracy of their answering decisions.

[0268] Through this implementation, the functionality and information transparency of incoming call reminders can be significantly improved. By displaying additional call information or call intention about the calling number on the incoming call reminder interface and / or the incoming call pre-reminder interface, the called user can obtain more information about the caller in advance, helping the called user to quickly understand the background and purpose of the call, thereby improving the accuracy of the answering decision and helping to make more informed answering decisions; at the same time, it can reduce the confusion caused by unclear call background to users, thereby improving communication efficiency; in addition, this implementation can ensure that users obtain sufficient information support before answering the call, avoiding misunderstandings or delays caused by information asymmetry.

[0269] In some modified implementations of the embodiments of the present application, the method may further include:

[0270] In response to a contact adding operation for the calling number, obtaining an electronic business card corresponding to the calling number;

[0271] The calling user is added as a contact according to the identity information recorded in the electronic business card.

[0272] Among them, the contact adding operation refers to the action performed by the user on the communication application or device, which is intended to save a certain phone number (in this case, the calling number) to the contact list, thereby adding the phone number as a contact. An electronic business card is a digital personal information card that usually contains the identity information of the calling user. Identity information refers to basic information used to identify the calling user, including but not limited to name, company name, contact information, etc. In addition, the electronic business card provided in the embodiment of the present application can also record business information related to the current call (such as the purpose of the call, appointment matters, etc.). The electronic business card can be sent directly to the called terminal through the application or communication protocol, or it can be uploaded to the server for storage. When the user performs the operation of adding the calling number as a contact, the called terminal can automatically detect and obtain the electronic business card corresponding to the calling number, and complete the contact adding process based on the user information recorded in the electronic business card. Among them, the electronic business card can be obtained in real time from the calling terminal or the server, or it can be obtained in advance and called in real time. Since the electronic business card is edited by the calling user, the identity information recorded in it is more reliable, so the identification of the incoming call intention is more accurate. In addition, the electronic business card can also be displayed on the incoming call pre-alert interface or the incoming call reminder interface to help the called user better understand the background of the call and make more informed answering decisions.

[0273] In some examples, when the called user performs a contact addition operation to add the caller number as a contact, the called terminal can immediately start a background process to find and obtain the electronic business card associated with the caller number. This electronic business card can be stored in the calling terminal, a cloud server, a local database, or other accessible data sources. Once the electronic business card is obtained, the user information in it can be parsed and automatically filled into the relevant fields of the contact addition interface, such as name, phone number, email address, etc., and then a new contact entry is created based on this information and saved to the user's contact list.

[0274] This implementation significantly enhances the intelligence of contact management. First, by automatically acquiring and utilizing electronic business card information, the need for users to manually enter contact information is reduced, thereby streamlining operational processes and improving efficiency. Second, the accuracy and completeness of contact information can be ensured, avoiding information inconsistencies caused by manual entry errors. Furthermore, since the electronic business card originates from the caller and is initially generated by the caller, the accuracy and authenticity of the electronic business card information can be ensured. Furthermore, this automated processing enhances the user experience, allowing users to complete multiple tasks in a single operation, improving overall operational efficiency and user satisfaction.

[0275] In some application scenarios, the called terminal may be a device that supports multiple SIM (Subscriber Identity Module, User Identity Module) cards, such as a dual-card device. A dual-card device refers to a mobile device with two built-in SIM card slots, allowing the user to insert two different SIM cards at the same time, so that two independent phone numbers and data connections can be managed on the same mobile phone. In this case, the same call processing configuration (such as intelligent interaction conditions, first harassment call determination conditions, and second harassment call determination conditions, etc.) can be set for different phone numbers of the called terminal, or different call processing configurations can be set for phone numbers corresponding to different SIMs, and the calls can be differentiated and processed according to the called number. The above-mentioned call processing method can first determine the called number requested by the incoming call request, and execute the above-mentioned call processing method according to the call processing configuration corresponding to the called number requested to call. Therefore, the call processing method, intelligent interaction conditions, paid answering process, settings and specific steps introduced in the above embodiments can also be for the called number, which will not be repeated here one by one, and should also be within the scope of protection of this application.

[0276] It should be noted that the call processing method based on paid answering provided in this embodiment can coexist with the existing harassment call blocking solution and be implemented in combination with each other to obtain a more personalized anti-harassment function, meet the diverse usage needs of users, and provide users with a more intelligent and flexible communication environment. For example, the existing commonly used harassment call blocking solution usually intercepts harassment calls based on number tag information. After integrating the technical solution of this application on this basis, it is possible to directly intercept strange calls of some number categories based on the number tag information, while executing artificial intelligence voice interaction and paid answering processes for strange calls of other number categories. For example, for phone numbers that have been repeatedly marked as advertising or fraud, they can be directly intercepted to prevent these unwanted calls from disturbing the user. For those cases where they are not widely marked but are still strange numbers, it is possible to choose to execute artificial intelligence voice interaction and paid answering processes to ensure that only calls willing to pay can trigger call reminders on the called terminal. In addition, it is also possible to determine whether to directly intercept, trigger artificial intelligence voice interaction, or directly execute the paid answering process based on the number of times the phone number is marked. For example, if the number of marks is less than the first preset threshold, the paid answering process is directly executed; if the number of marks is greater than the first preset threshold and less than the second preset threshold, artificial intelligence voice interaction is triggered; if the number of marks is greater than the second preset threshold, the call is directly intercepted, thereby effectively screening out high-risk harassing calls. Among them, the first preset threshold and the second preset threshold can be flexibly set according to user needs, or directly set by the user, and the embodiments of this application are not limited.

[0277] By organically combining artificial intelligence voice interaction, paid answering solutions and harassment call blocking solutions, users can be provided with more personalized and secure anti-harassment services, meeting diverse usage needs, thereby improving the user's call experience. In addition, by setting a reasonable marking threshold, the processing strategy for different calls can be dynamically adjusted, making the entire anti-harassment system more intelligent and efficient.

[0278] It should be noted that in the description of the embodiments of the present application, the part executed by the called terminal or the call application can be directly executed by the call application or the auxiliary application, while the part executed by the server can be triggered by the called terminal, the call application or the auxiliary application sending relevant data and requests to the server to trigger the server execution, and obtain the execution result data from the server. The relevant points can be understood by referring to the description of this part. They are all simple changes to the implementation method of the present application and should all be within the scope of protection of the present application.

[0279] In some embodiments, the aforementioned number category can be determined based on number classification information, which can be determined based on at least one of number tag information, number registration information, and number authentication information. Number tag information refers to information assigned to a number by a third-party service provider or user, and can include information such as courier phone, sales phone, or fraud phone. Number registration information can refer to the identity information selected by the number holder when registering on the service platform corresponding to the first call application, or the identity information selected for the number on the service platform. A user can register different identity information for different numbers they own, or they can register the same identity information. For example, a courier with two phone numbers can register phone number A, used for deliveries, as a courier phone, and phone number B, used for personal use, as a private phone, or they can omit identity registration. Number authentication information can refer to the identity information determined by the number holder after authenticating the number on the service platform corresponding to the first call application. Unlike number registration information, number authentication information requires the number holder to provide necessary identification documents for successful authentication. Therefore, number authentication information is more authentic and reliable than identity registration information, helping to more accurately determine the number category of a phone number.

[0280] The first call application and the service end (for example, the service platform or server corresponding to the first call application) can determine the number classification information of the telephone number (for example, the calling number) based on at least one of the above number marking information, number registration information and number authentication information. For example, if the number marking information of a certain telephone number is a fraud call, its number classification information can be determined as a fraud call; if the number authentication information of a certain telephone number is an express call, its number classification information can be determined as an express call; if the number marking information, number registration information and number authentication information of a certain telephone number are all empty values, that is, there is no corresponding number marking information, number registration information and number authentication information, then its number classification information can be determined as an unclassified call, thereby realizing the classification of telephone numbers.

[0281] Considering that the number tag information, number registration information, and number authentication information corresponding to the same phone number may be different, which brings confusion to the classification of phone numbers, therefore, in some implementations of the embodiments of the present application, different priorities can be set for the number tag information, number registration information, and number authentication information according to the preset number classification strategy, and then the number classification information of the phone number (such as the calling number) is determined based on the above number tag information, number registration information, and number authentication information, wherein the number authentication information has the highest priority, followed by the number tag information, and finally the number registration information. When there are multiple different information for the same phone number, the information with the highest priority is used for classification. For example, if the number authentication information of a phone number is an express call, even if its number tag information is a marketing call, it will still be classified as an express call; for another example, if the number registration information of a phone number is an express call, but its number tag information is a fraud call, and its number authentication information is empty, it will be classified as a fraud call.

[0282] Furthermore, these priorities can be customized by the user. For example, users can customize the priority of each information source within the paid call conditions editing interface. For example, a user can prioritize number tagging information, ensuring that the system prioritizes number tagging information from third-party service providers or users. This customized priority setting can meet user needs and enhance system flexibility.

[0283] The above number classification strategy can more accurately determine the category of a phone number, thereby helping users to manage incoming calls more finely, ensuring that important calls can be connected smoothly, while reducing unnecessary harassing calls.

[0284] Based on the above embodiments, the call processing method provided in the embodiments of the present application can at least achieve the following technical effects:

[0285] 1. Raising the threshold for answering incoming calls: By triggering a paid answering process for harassing calls, the caller of a harassing call needs to pay a fee to trigger a call reminder on the called terminal. If the caller is unwilling to pay, the call reminder will not be triggered. Since callers are generally unwilling to pay additional fees, this can effectively reduce the number of call reminders and actual call volume for harassing calls, intercept some harassing calls, and reduce the negative impact of harassing calls on called users from the perspective of raising the answering threshold and reducing the number of harassing calls.

[0286] 2. Generates income for the called user: Since answering a harassing call requires the calling user to pay a certain fee to the called user, if the calling user is willing to pay the fee to trigger the incoming call reminder, the called user can receive a certain amount of financial compensation when answering the harassing call, which brings additional income to the called user. This financial compensation can effectively alleviate the negative emotions of the called user when answering harassing calls, thereby reducing the negative impact of harassing calls on the called user from the perspective of financial compensation.

[0287] 3. Improve the flexibility of call processing: This application solution has a high degree of flexibility in intercepting harassing calls. It does not simply intercept calls based on number marking information. Even if the number is marked as a harassing call, if the calling user is willing to pay the answering fee, the call can still be successfully processed, thus avoiding the situation where important calls are mistakenly intercepted and ensuring that users do not miss truly important calls.

[0288] 4. Enhanced user control: This application solution supports users to set personalized intelligent interaction conditions, first and second harassing call determination conditions, charging standards, etc. according to their own needs. This flexible setting method enables users to better control the call environment, reduce unnecessary interference, and obtain more suitable financial compensation, thereby improving the user experience.

[0289] 5. Improve the intelligence level of call processing: Through the artificial intelligence voice interaction process, it can more intelligently identify nuisance calls, ensuring that only nuisance calls trigger the paid answering process, thereby improving the intelligence level of call processing and managing and handling calls more accurately.

[0290] 6. Reduce user time and energy waste: Since callers need to pay to trigger call alerts, and most callers only pay for important or valuable calls, users can more confidently confirm the other party's true intentions before answering the call from a caller number that can trigger a call alert, thereby reducing users' time and energy wasted on answering nuisance calls and improving call efficiency. By paying to answer calls, users can answer calls with greater confidence because they know these calls have been screened and confirmed. This not only reduces user anxiety and uneasiness, but also improves the overall user experience.

[0291] 7. Enhanced protection of personal information: The paid answering mechanism can effectively prevent the called user from being harassed by arbitrary calls, especially against malicious phone bombing. The called user can choose to only answer calls that are willing to pay, thereby protecting their personal information from infringement. Even if they are bombarded with calls, they can still receive certain financial compensation.

[0292] 8. Benefits to the calling user: The call processing method provided by the embodiment of the present application is not only beneficial to the called user, but also has many benefits for the calling user. For example, taking the calling user engaged in marketing business as an example, this solution can improve the connection rate and conversion rate of marketing calls, and help the called user screen out high-quality marketing calls. Since the initiator of the marketing call usually hopes to establish contact with potential customers through the phone, paying the fee can serve as a screening mechanism to ensure that only truly sincere marketing calls can be successfully dialed, which can not only improve the connection rate of marketing calls, but also improve the effectiveness of marketing activities. For the initiator of the marketing call, paying the fee can increase the value of the call, which not only shows the sincerity of the marketing call, but also increases the willingness of the called user to answer the call. When the called user knows that the other party has paid the fee, he is more likely to answer the call, thereby improving the connection rate of the marketing call. When the called users know that the initiator of the marketing call has paid the fee, they will have a higher sense of trust in these calls. This sense of trust can make the called users more willing to answer the calls, thereby improving the connection rate and conversion rate of marketing calls. Therefore, the embodiment of the present application can also promote the healthy development of telephone sales business by charging for answering incoming calls, reduce or eliminate users' resistance and rejection of telephone sales, and enable users who really need it to get better quality services.

[0293] In the above embodiment, a method for handling incoming calls is provided. Correspondingly, the present application also provides a device for handling incoming calls. The device for handling incoming calls provided in the embodiment of the present application can implement the above method for handling incoming calls. The device for handling incoming calls can be implemented by software, hardware, or a combination of software and hardware. For example, the device for handling incoming calls can include integrated or separate functional modules or units to perform the corresponding steps in the above methods. Please refer to Figure 3 , which schematically shows a schematic diagram of a call processing device provided by some embodiments of the present application. Since the device embodiment is basically similar to the method embodiment, the description is relatively simple. For relevant parts, please refer to the partial description of the method embodiment. The device embodiment described below is only illustrative. Figure 3 As shown, the incoming call processing device may include:

[0294] The artificial intelligence voice interaction module 101 is used to connect the incoming call and conduct artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant when detecting that the incoming call meets the preset intelligent interaction conditions;

[0295] The harassing call determination module 102 is used to determine whether the incoming call is a harassing call based on the artificial intelligence voice interaction content;

[0296] The paid answering execution module 103 is used to execute a paid answering process for the incoming call determined to be a harassing call. The paid answering process includes an incoming call reminder step, and the incoming call reminder step includes triggering an incoming call reminder after detecting payment confirmation information corresponding to the calling number.

[0297] In some modified implementations, the harassing call determination module 102 includes:

[0298] A call intention determination unit, configured to determine the call intention of the calling user based on the artificial intelligence voice interaction content;

[0299] The first harassing call determination unit is configured to determine that the incoming call is a harassing call if the incoming call intention meets a first harassing call determination condition, wherein the first harassing call determination condition is configured to determine that the incoming call is a harassing call based on the incoming call intention.

[0300] In some implementation variations, the incoming call intention determination unit includes:

[0301] a semantic recognition subunit, configured to perform semantic recognition on the artificial intelligence voice interaction content according to a semantic recognition model, and determine the calling user's calling intention according to the semantic recognition result; and / or,

[0302] The keyword matching subunit is used to perform keyword matching on the artificial intelligence voice interaction content according to the preset call intention keywords, and determine the call intention of the calling user based on the matching results.

[0303] In some modified implementations, the harassing call determination module 102 includes:

[0304] A second harassing call judgment unit is configured to perform keyword matching on the artificial intelligence voice interaction content according to preset harassing call keywords, and judge whether the incoming call is a harassing call according to the matching result; or

[0305] The third harassing call judgment unit is used to input the artificial intelligence voice interaction content into a pre-trained harassing call judgment model, and judge whether the incoming call is a harassing call according to the output result of the harassing call judgment model.

[0306] In some variations, the device further comprises:

[0307] The paid answering direct trigger module is used to trigger the paid answering process when it is detected that the incoming call meets the second harassing call determination condition, and the second harassing call determination condition is used to determine whether the incoming call is a harassing call based on the incoming call information.

[0308] In some variations, the device further comprises:

[0309] A contact editing interface display module, used to display a paid answering indicator setting option in the contact editing interface for the calling number;

[0310] The paid answering mark setting module is used to set a paid answering mark for the calling number according to the selection operation of the called user on the paid answering mark setting option, and the paid answering mark is used to mark that the calling number meets the second harassing call determination condition.

[0311] In some modified implementations, the artificial intelligence voice interaction module 101 includes:

[0312] The silent interaction unit is used to answer the incoming call while keeping the incoming call silent at the front desk of the called terminal and to perform artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant.

[0313] In some modified implementations, the paid call execution module 103 includes:

[0314] The paid answering information interaction unit is used to interact with the calling user through an artificial intelligence voice assistant to exchange paid answering information, wherein the paid answering information includes at least one of paid answering prompt information, payment guidance information, payment timing option information and automatic payment option information.

[0315] In some variations, the device further comprises:

[0316] The incoming call pre-alert module is used to trigger an incoming call pre-alert for an incoming call when it is detected that the pre-alert conditions are currently met. The incoming call pre-alert includes displaying an incoming call pre-alert interface, and the incoming call pre-alert interface includes at least one of an answer trigger control, a reject trigger control, a paid answer trigger control, and an artificial intelligence agent trigger control.

[0317] In some variations, the device further comprises:

[0318] The incoming call connection module is used to connect the incoming call according to the direct answering operation input by the called user in response to the incoming call pre-alert;

[0319] A call rejection module is used to reject the incoming call according to the rejection operation input by the called user in response to the incoming call pre-alert;

[0320] A paid answering module is used to trigger the execution of a paid answering process according to the paid answering operation input by the called user in response to the incoming call pre-alert;

[0321] An intelligent answering module, configured to trigger the step of answering the incoming call and performing artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant according to the artificial intelligence answering operation input by the called user in response to the incoming call pre-alert;

[0322] The no-operation processing module is used to close the incoming call pre-alert according to detecting that there is no operation on the incoming call pre-alert within a preset time.

[0323] In some modified implementations, the payment confirmation information includes committed payment information; accordingly, the apparatus further includes:

[0324] A breach of contract processing module is used to execute breach of contract processing steps according to the promised payment information after the call ends, and the breach of contract processing steps include at least one of the following: when it is detected that the promised payment information has not been fulfilled within a first time limit, sending a call fee recovery reminder message to the calling terminal, and the call fee recovery reminder message is used to prompt the calling user to pay the call fee for this call; when it is detected that the promised payment information has not been fulfilled within a second time limit, sending a breach of trust information for the calling number to the service end, and the breach of trust information is used for other called terminals connected to the server end to restrict the calling number from answering calls.

[0325] In some variations, the device further comprises:

[0326] Lost signal code detection module, used to detect whether the calling number is a lost signal code;

[0327] The lost signal code restriction module is used to restrict the calling number from answering calls when it is detected that the calling number is a lost signal code.

[0328] In some modified implementations, the paid answering execution module 103 further includes:

[0329] The automatic payment unit is used to send an automatic payment request to the server, wherein the automatic payment request is used to request the server to automatically pay the answering fee from the payment account bound to the calling number according to the automatic payment agreement corresponding to the calling number.

[0330] In some variations, the device further comprises:

[0331] The answering time protection module is used to execute the answering time protection process after detecting that the called user answers the call. The answering time protection process includes using a preset answering time protection method to protect the answering time of this call.

[0332] In some implementation variations, the call duration protection module includes:

[0333] The condition-compliant protection unit is used to execute the answering time protection process after detecting that the called user answers the call when the calling number meets the answering time protection condition.

[0334] In some variations, the device further comprises:

[0335] The refund processing module is used to trigger the execution of the refund processing process when the refund trigger conditions are met. The refund processing process includes fully or partially refunding the answering fee paid by the calling user.

[0336] In some variations, the device further comprises:

[0337] A manual refund element display module is used to display a call end interface after the call ends, and to display a manual refund trigger element on the call end interface;

[0338] The manual refund triggering module is used to trigger the execution of the refund processing flow in response to the manual refund operation input by the called user based on the manual refund triggering element.

[0339] In some implementation variations, the refund processing module includes:

[0340] The refund unit is used to refund the call fee paid by the calling user according to the refund amount input or confirmed by the called user.

[0341] In some variations, the device further comprises:

[0342] The additional information display module is used to display additional call information and / or call intention about the calling number on the call reminder interface and / or call pre-alert interface. The additional call information includes the identity information of the calling user and / or call description information for this call.

[0343] In some variations, the device further comprises:

[0344] An electronic business card acquisition module, configured to acquire an electronic business card corresponding to a calling number in response to a contact adding operation for the calling number;

[0345] The contact adding module is used to add the calling user as a contact according to the identity information recorded in the electronic business card.

[0346] In some modified implementations, the payment confirmation information includes confirmation information that the calling user pays the called user a call fee, and the call fee is determined according to a charging standard preset by the called terminal; the apparatus further includes:

[0347] Charging standard editing element display module, used to display charging standard editing elements in the charging standard setting interface;

[0348] The charging standard editing module is used to determine the charging standard according to the editing operation of the called user on the charging standard editing element.

[0349] In some variations, the device further comprises:

[0350] The intention processing module is used to select a call processing method corresponding to the call intention determined based on the artificial intelligence voice interaction content to handle the call.

[0351] In some modified implementations, the artificial intelligence voice interaction module 101 includes:

[0352] The background connection unit is used to connect the incoming call in the background and conduct artificial intelligence voice interaction with the calling user through the artificial intelligence voice assistant. The background connection of the incoming call includes connecting the incoming call while disconnecting the external voice collection link and call playback link of the called terminal.

[0353] The incoming call processing device provided in the embodiment of the present application is based on the same inventive concept as the incoming call processing method provided in the aforementioned embodiment of the present application and has the same beneficial effects, which will not be described in detail here.

[0354] The embodiment of the present application also provides an incoming call processing system corresponding to the incoming call processing method provided in the above embodiment, comprising: a called terminal and a server; wherein the called terminal is in communication connection with the server; the called terminal is used to execute the incoming call processing method provided in the embodiment of the present application; the server is used to provide service support for the called terminal to execute the incoming call processing method. Figure 1 The system framework diagram shown and the aforementioned embodiment description of the call processing method can be understood. The call processing system provided by the embodiment of the present application and the call processing method provided by the aforementioned embodiment of the present application are based on the same inventive concept and have the same beneficial effects, which will not be repeated here.

[0355] An embodiment of the present application also provides an electronic device corresponding to the incoming call processing method provided in the aforementioned embodiment. The electronic device can be any device with communication capabilities, such as a mobile phone, a landline phone (landline), a smart watch, a smart bracelet, a children's phone watch, a tablet computer, a vehicle-mounted terminal (smart car machine), smart glasses, etc., to execute the above-mentioned incoming call processing method.

[0356] Please refer to Figure 4 , which schematically shows a schematic diagram of an electronic device provided by some embodiments of the present application. Figure 4As shown, the electronic device 40 includes: a processor 400, a memory 401, a bus 402 and a communication interface 403, and the processor 400, the communication interface 403 and the memory 401 are connected via the bus 402; the memory 401 stores a computer program that can be run on the processor 400, and when the processor 400 runs the computer program, it executes the call processing method provided in any of the aforementioned embodiments of the present application.

[0357] Among them, the memory 401 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 403 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 402 can be an ISA bus, a PCI bus or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. Among them, the memory 401 is used to store programs, and the processor 400 executes the program after receiving the execution instruction. The incoming call processing method disclosed in any implementation method of the aforementioned embodiment of the present application can be applied to the processor 400, or implemented by the processor 400.

[0358] The processor 400 may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor 400 or by software instructions. The above processor 400 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. It can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 401 , and the processor 400 reads the information in the memory 401 and completes the steps of the above method in combination with its hardware.

[0359] The electronic device provided in the embodiment of the present application and the incoming call processing method provided in the aforementioned embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the methods adopted, operated or implemented therein.

[0360] The present application also provides a computer-readable medium corresponding to the call processing method provided in the above embodiment. Figure 5 The computer-readable storage medium shown is a CD 50 on which a computer program (ie, a program product) is stored. When the computer program is run by a processor, the incoming call processing method provided by any of the aforementioned embodiments will be executed.

[0361] It should be noted that examples of the computer-readable storage medium may also include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other optical or magnetic storage media, which are not listed here one by one.

[0362] The computer-readable storage medium provided in the above-mentioned embodiment of the present application and the incoming call processing method provided in the above-mentioned embodiment of the present application are based on the same inventive concept and have the same beneficial effects as the method adopted, run or implemented by the application program stored therein.

[0363] It should be noted that the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present application. Among them, each box in the flowchart or block diagram can represent a module, program segment or part of the code, and the module, program segment or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0364] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0365] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0366] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0367] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0368] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0369] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and they should all be included in the scope of the claims and description of the present application.

Claims

1. A method for processing incoming calls, applied to a called terminal, characterized in that: include: When it is detected that the incoming call meets the preset intelligent interaction conditions, the call is answered and the caller is interacted with through the artificial intelligence voice assistant; Determine whether the incoming call is a nuisance call based on the content of the artificial intelligence voice interaction; A paid answering process is executed for the incoming call determined to be a harassing call, and the paid answering process includes an incoming call reminder step, and the incoming call reminder step includes triggering the incoming call reminder after detecting the payment confirmation information corresponding to the calling number.

2. The method according to claim 1, characterized in that The intelligent interaction condition includes at least one of the following: The calling number is an unfamiliar number; The calling number is an unfamiliar number and has no number tag; The calling number is an unfamiliar number and has a designated first number mark for artificial intelligence voice interaction; The calling number is an unfamiliar number and the calling user identity or calling purpose indicated by the corresponding additional call information requires artificial intelligence voice interaction. The additional call information includes the calling user's identity information and / or call description information; The calling number is an unfamiliar number and its number category falls under the preset situation where artificial intelligence voice interaction is required; The incoming call time falls within the preset intelligent interaction period; The AI ​​voice interaction is manually triggered by the called user.

3. The method according to claim 1, characterized in that The determining whether the incoming call is a nuisance call based on the artificial intelligence voice interaction content includes: Determining the calling intention of the calling user based on the artificial intelligence voice interaction content; If the incoming call intention meets the first harassing call determination condition, the incoming call is determined to be a harassing call, and the first harassing call determination condition is used to determine whether the incoming call is a harassing call based on the incoming call intention.

4. The method according to claim 3, characterized in that The determining the calling user's calling intention based on the artificial intelligence voice interaction content includes: Performing semantic recognition on the artificial intelligence voice interaction content according to the semantic recognition model, and determining the calling user's calling intention according to the semantic recognition result; and / or, The artificial intelligence voice interaction content is keyword matched according to the preset call intention keywords, and the call intention of the calling user is determined based on the matching results.

5. The method according to claim 1, wherein The determining whether the incoming call is a nuisance call based on the artificial intelligence voice interaction content includes: Perform keyword matching on the AI ​​voice interaction content based on the preset harassing call keywords, and determine whether the incoming call is a harassing call based on the matching results; or The artificial intelligence voice interaction content is input into a pre-trained harassing call judgment model, and whether the incoming call is a harassing call is determined based on the output result of the harassing call judgment model.

6. The method according to claim 1, characterized in that The paid answering process further includes: a paid answering information interaction step; The paid answering information interaction step includes interacting with the calling user through an artificial intelligence voice assistant to exchange paid answering information, and the paid answering information includes at least one of paid answering prompt information, payment guidance information, payment timing option information and automatic payment option information.

7. The method according to claim 1, characterized in that The method further comprises: If the refund triggering conditions are met, a refund processing procedure is triggered, which includes a full or partial refund of the call fee paid by the calling user; The refund triggering condition includes at least one of the following: the call is not answered by the called user, the call is ended by the called user and the answering time is less than the preset answering protection time, and a manual refund operation input by the called user is detected.

8. A call processing device, applied to a called terminal, characterized in that: include: The artificial intelligence voice interaction module is used to connect the incoming call and conduct artificial intelligence voice interaction with the caller through the artificial intelligence voice assistant when it detects that the incoming call meets the preset intelligent interaction conditions; A harassing call determination module is used to determine whether the incoming call is a harassing call based on the content of the artificial intelligence voice interaction; The paid answering execution module is used to execute a paid answering process for the incoming call determined to be a harassing call. The paid answering process includes an incoming call reminder step, and the incoming call reminder step includes triggering the incoming call reminder after detecting the payment confirmation information corresponding to the calling number.

9. A call processing system comprising: The called terminal and the server; The called terminal is in communication with the service end; The called terminal is used to execute the incoming call processing method according to any one of claims 1 to 7; The service end is used to provide service support for the called terminal to execute the incoming call processing method.

10. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method according to any one of claims 1 to 7.

11. A computer-readable storage medium, characterized in that Computer-readable instructions are stored thereon, and the computer-readable instructions can be executed by a processor to implement the method according to any one of claims 1 to 7.

Images