Call processing method, device, electronic device and storage medium

Abstract

The present disclosure provides a call processing method, device, electronic device and storage medium, which belongs to the field of communication technology. In the embodiment of the present disclosure, in the target call event triggered by the historical call event of the target object, the call mode of the voice robot is first used to talk to the target object. In this way, when facing a large number of telecommunications network frauds, the call mode of the voice robot is preferentially used, which can effectively reduce the call pressure and improve the call efficiency. During the call, based on the call audio between the voice robot and the target object, the safety reference value of the target object in the historical call event is determined. When the safety reference value does not meet the safety conditions, the manual call mode is used to talk to the target object to ensure the call efficiency. In this way, a human-machine combined call dissuasion method is provided, which can improve the dissuasion efficiency while ensuring the dissuasion effect.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a call processing method, device, electronic device and storage medium. Background Art

[0002] In recent years, the rapid development of information technology has brought many benefits to people, but also triggered some new risks, such as telecommunications network fraud. When facing the problem of telecommunications network fraud, dissuading the victim has become an important means to prevent such problems. Among them, dissuasion means that when the victim receives a suspected fraudulent call, the victim is dissuaded by calling the victim and talking to him / her, so as to avoid the risk of being defrauded.

[0003] At present, due to the large amount of telecommunications network fraud data, there are too many victims who need to be dissuaded. Therefore, how to improve the efficiency of dissuasion while ensuring the effectiveness of dissuasion is an urgent problem that needs to be solved. Summary of the invention

[0004] The disclosed embodiments provide a call processing method, device, electronic device and storage medium, which can improve the dissuasion efficiency and ensure the dissuasion effect. The technical solution includes the following steps.

[0005] In one aspect, a call processing method is provided, the method comprising:

[0006] In a target call event triggered by a historical call event of the target object, controlling the voice robot to talk to the target object, the target call event is used to provide a safety prompt to the target object for the historical call event;

[0007] Determine a safety reference value of a historical call event of the target object based on a first call audio of the call between the voice robot and the target object;

[0008] If the safety reference value does not meet the safety conditions, switch to manual call mode.

[0009] In some embodiments, based on the first call audio of the call between the voice robot and the target object, determining the safety reference value of the historical call event of the target object includes:

[0010] Extracting call content information and / or emotion indication information of the target object from the first call audio, where the emotion indication information represents the emotional characteristics of the target object in the target call event;

[0011] Based on the call content information and / or emotion indication information of the target object, a safety reference value of the historical call events of the target object is determined.

[0012] In some embodiments, based on the call content information and the emotion indication information of the target object, determining the safety reference value of the historical call events of the target object includes:

[0013] The call content information and emotion indication information of the target object are input into a security recognition model, and the security reference value of the historical call events of the target object is predicted by the security recognition model to obtain the security reference value of the historical call events of the target object. The security recognition model is trained based on the call content information, emotion indication information and security reference value of multiple sample objects.

[0014] In some embodiments, after determining that the safety reference value does not meet the safety condition and before switching to the manual call mode, the method further includes:

[0015] The call information of the target call event is sent to the management terminal, where the call information includes the call content information of the first call audio.

[0016] In some embodiments, the call information further includes state information of the target object, the state information indicating the state of the target object in the target call event;

[0017] After determining the safety reference value of the target object's historical call events based on the first call audio of the voice robot and the target object, the method further includes:

[0018] If the safety reference value does not meet the safety condition, the status information of the target object is determined based on the first call audio.

[0019] In some embodiments, after determining that the safety reference value does not meet the safety condition and before switching to the manual call mode, the method further includes:

[0020] The voice robot is controlled to play target audio data, where the target audio data is used to prompt that the manual call mode is about to be used for the call.

[0021] In some embodiments, after switching to the manual call mode, the method further includes:

[0022] Semantic recognition is performed on the call content information of the second call audio of the manual call mode to obtain call effect information of the second call audio, where the call effect information indicates the effect of the security prompt on the target call event.

[0023] In some embodiments, after switching to the manual call mode, the method further includes:

[0024] Extracting call content information corresponding to the manual call mode from the second call audio of the manual call mode;

[0025] A quality reference value of the call content information corresponding to the manual call mode is determined, where the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode.

[0026] In some embodiments, determining the quality reference value of the call content information corresponding to the manual call mode includes at least one of the following:

[0027] Determine a quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event, wherein the call effect information indicates the effect of the security prompt of the target call event;

[0028] Determining a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event;

[0029] Determining a quality reference value of the call content information corresponding to the manual call mode based on a second matching degree between the call content information corresponding to the manual call mode and the historical call event;

[0030] Based on the rarity of the call content information corresponding to the manual call mode, a quality reference value of the call content information corresponding to the manual call mode is determined, where the rarity indicates the rarity of the corresponding call content information.

[0031] In some embodiments, after determining the quality reference value of the call content information corresponding to the manual call mode, the method further includes:

[0032] If the quality reference value meets the content quality condition, the content information library is updated based on the call content information corresponding to the manual call mode, and the content information library includes a plurality of content information for providing safety prompts.

[0033] In one aspect, a call processing device is provided, the device comprising:

[0034] A call module, used to control the voice robot to talk to the target object in a target call event triggered by a historical call event of the target object, wherein the target call event is used to provide a safety prompt to the target object for the historical call event;

[0035] A determination module, configured to determine a safety reference value of a historical call event of the target object based on a first call audio of a call between the voice robot and the target object;

[0036] The switching module is used to switch to the manual call mode if the safety reference value does not meet the safety condition.

[0037] In some embodiments, the determining module includes:

[0038] An extraction submodule, configured to extract call content information and / or emotion indication information of the target object from the first call audio, wherein the emotion indication information represents an emotional feature of the target object in the target call event;

[0039] The determination submodule is used to determine the safety reference value of the historical call events of the target object based on the call content information and / or emotion indication information of the target object.

[0040] In some embodiments, a submodule is determined to:

[0041] The call content information and emotion indication information of the target object are input into a security recognition model, and the security reference value of the historical call events of the target object is predicted by the security recognition model to obtain the security reference value of the historical call events of the target object. The security recognition model is trained based on the call content information, emotion indication information and security reference value of multiple sample objects.

[0042] In some embodiments, the apparatus further comprises:

[0043] The sending module is used to send the call information of the target call event to the management terminal, and the call information includes the call content information of the first call audio.

[0044] In some embodiments, the call information further includes state information of the target object, the state information indicating the state of the target object in the target call event;

[0045] The determination module is also used for:

[0046] If the safety reference value does not meet the safety condition, the status information of the target object is determined based on the first call audio.

[0047] In some embodiments, the apparatus further comprises:

[0048] The control module is used to control the voice robot to play target audio data, and the target audio data is used to prompt that the manual call mode is about to be used for the call.

[0049] In some embodiments, the apparatus further comprises:

[0050] The identification module is used to perform semantic recognition on the call content information of the second call audio of the manual call mode to obtain the call effect information of the second call audio, and the call effect information indicates the effect of the security prompt of the target call event.

[0051] In some embodiments, the device further includes an extraction module for extracting call content information corresponding to the manual call mode from the second call audio of the manual call mode;

[0052] The determination module is further used to determine a quality reference value of the call content information corresponding to the manual call mode, and the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode.

[0053] In some embodiments, the determination module is further configured to:

[0054] Determine a quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event, wherein the call effect information indicates the effect of the safety prompt of the target call event;

[0055] Determining a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event;

[0056] Determining a quality reference value of the call content information corresponding to the manual call mode based on a second matching degree between the call content information corresponding to the manual call mode and the historical call event;

[0057] Based on the rarity of the call content information corresponding to the manual call mode, a quality reference value of the call content information corresponding to the manual call mode is determined, where the rarity indicates the rarity of the corresponding call content information.

[0058] In some embodiments, the apparatus further comprises:

[0059] The updating module is used to update the content information library based on the call content information corresponding to the manual call mode if the quality reference value meets the content quality condition. The content information library includes multiple content information for security prompts.

[0060] On the one hand, an electronic device is provided, which includes one or more processors and one or more memories, wherein the one or more memories store at least one program code, and the program code is loaded and executed by the one or more processors to implement the call processing method.

[0061] On the one hand, a computer-readable storage medium is provided, in which at least one program code is stored, and the program code is loaded and executed by a processor to implement the call processing method.

[0062] On the one hand, a computer program product is provided, which includes a computer program code, the computer program code is stored in a computer-readable storage medium, a processor of an electronic device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, so that the electronic device performs the above-mentioned call processing method.

[0063] The technical solution provided by the embodiment of the present disclosure first adopts the call mode of a voice robot to talk to the target object in a target call event triggered by the historical call event of the target object. In this way, when facing a large number of telecommunications network frauds, the call mode of the voice robot is preferentially adopted, which can effectively reduce the call pressure and improve the call efficiency. During the call, based on the call audio between the voice robot and the target object, the safety reference value of the target object in the historical call event is determined. When the safety reference value does not meet the safety conditions, the manual call mode is adopted to talk to the target object to ensure the call efficiency. In this way, a human-machine combined call dissuasion method is provided, which can improve the dissuasion efficiency while ensuring the dissuasion effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0065] Figure 1 It is a schematic diagram of an implementation environment of a call processing method provided by an embodiment of the present disclosure;

[0066] Figure 2 is a flow chart of a call processing method provided by an embodiment of the present disclosure;

[0067] Figure 3 is a flow chart of a call processing method provided by an embodiment of the present disclosure;

[0068] Figure 4 is a schematic diagram of a call management interface provided by an embodiment of the present disclosure;

[0069] Figure 5 is a structural diagram of a call processing device provided by an embodiment of the present disclosure;

[0070] Figure 6 is a schematic diagram of the structure of a terminal provided by an embodiment of the present disclosure;

[0071] Figure 7 It is a structural diagram of a server provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0072] In order to make the objectives, technical solutions and advantages of the present disclosure more clear, the embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.

[0073] The call processing method involved in the embodiments of the present disclosure can be applied to the scenario of telecommunication network fraud. In some embodiments, when the victim receives a suspected fraud call, the call processing method provided by the embodiments of the present disclosure can be used to initiate a call to the victim and talk to the victim to dissuade the victim, thereby avoiding the risk of being defrauded.

[0074] Figure 1 is a schematic diagram of an implementation environment of a call processing method provided by an embodiment of the present disclosure, see Figure 1 , the implementation environment includes: a first electronic device 101.

[0075] The first electronic device 101 may be provided as a server, and specifically may be a background server of a communication management platform, and the communication management platform provides functions of monitoring call events and initiating external calls.

[0076] The server can be an independent physical server, or a server cluster or distributed file system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. Optionally, the number of the above servers can be more or less, and the embodiments of the present disclosure are not limited to this. Of course, the server can also include other functional servers to provide more comprehensive and diversified services.

[0077] In the embodiment of the present disclosure, the first electronic device 101 is used to control a voice robot to communicate with a target object in a target call event triggered by a historical call event of the target object, and to determine a safety reference value of the historical call event of the target object based on a first call audio of the conversation between the voice robot and the target object; if the safety reference value does not meet the safety conditions, switching to a manual call mode.

[0078] In some embodiments, the implementation environment further includes: a second electronic device 102. The first electronic device 101 and the second electronic device 102 may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiments of the present disclosure.

[0079] The second electronic device 102 may be provided as a terminal, specifically a terminal operated by a manager (hereinafter referred to as a management terminal). In some embodiments, the second electronic device 102 is associated with a communication management platform. In the disclosed embodiment, the second electronic device 102 is used to provide a visual interface of the communication management platform for the manager to view and analyze.

[0080] In some embodiments, the terminal is at least one of a smart phone, a smart watch, a desktop computer, a laptop computer, a virtual reality terminal, an augmented reality terminal, a wireless terminal, and a laptop computer. The terminal may generally refer to one of a plurality of terminals, and this embodiment is only illustrated by a terminal. Those skilled in the art will appreciate that the number of the above terminals may be more or less.

[0081] Figure 2 is a flowchart of a call processing method provided by an embodiment of the present disclosure, see Figure 2 , performed by a first electronic device, the method includes the following steps.

[0082] 201. In a target call event triggered by a historical call event of the target object, the first electronic device controls a voice robot to talk to the target object, and the target call event is used to provide a safety prompt to the target object for the historical call event.

[0083] 202. The first electronic device determines a safety reference value of historical call events of the target object based on a first call audio of a call between the voice robot and the target object.

[0084] 203. If the safety reference value does not meet the safety condition, the first electronic device switches to a manual call mode.

[0085] The technical solution provided by the embodiment of the present disclosure first adopts the call mode of a voice robot to talk to the target object in a target call event triggered by the historical call event of the target object. In this way, when facing a large number of telecommunications network frauds, the call mode of the voice robot is preferentially adopted, which can effectively reduce the call pressure and improve the call efficiency. During the call, based on the call audio between the voice robot and the target object, the safety reference value of the target object in the historical call event is determined. When the safety reference value does not meet the safety conditions, the manual call mode is adopted to talk to the target object to ensure the call efficiency. In this way, a human-machine combined call dissuasion method is provided, which can improve the dissuasion efficiency while ensuring the dissuasion effect.

[0086] In the embodiment of the present disclosure, the first electronic device may be provided as a server, and the second electronic device may be provided as a management terminal. The process of call processing is described below with the server and the management terminal as the execution subjects. Figure 3is a flowchart of a call processing method provided by an embodiment of the present disclosure, see Figure 3 , the method comprises the following steps.

[0087] 301. In a target call event triggered by a historical call event of the target object, the server controls a voice robot to talk to the target object. The target call event is used to provide a security prompt to the target object for the historical call event.

[0088] The target object refers to the target user, which can be any user. In the embodiment of the present disclosure, the target user is used to refer to a user who has had an abnormal call event, and the abnormal call event refers to a call event with security issues, such as a fraudulent call. The embodiment of the present disclosure subsequently uses a historical call event to refer to the abnormal call event.

[0089] The target call event is triggered based on the historical call event of the target object, so as to provide a safety reminder to the target object for the historical call event. The safety reminder can be understood as a fraud warning or anti-fraud propaganda.

[0090] In some embodiments, the server provides a function of monitoring the user's call records. In an optional embodiment, if the server detects an abnormal call from the target object, a target call event is triggered, that is, a call is initiated to the target object to talk to the target object. For example, the abnormal call may be an overseas call.

[0091] 302. The server determines a safety reference value of the target object's historical call events based on the first call audio between the voice robot and the target object.

[0092] Among them, the first call audio is used to refer to the call audio between the voice robot and the target object. The security reference value is used to indicate the possibility of the existence of security problems in the historical call event. Understandably, taking telecommunications network fraud as an example, the security reference value is also the risk level of fraud risk. In some embodiments, the security reference value is expressed by the level of security problems, such as the fraud risk level; or, the security reference value is expressed by the probability of security problems, such as the probability value of being defrauded.

[0093] In some embodiments, the server extracts the call content information and / or emotion indication information of the target object from the first call audio, and determines the safety reference value of the historical call events of the target object based on the call content information and / or emotion indication information of the target object.

[0094] Among them, the call content information of the target object refers to the call content fed back by the target object in the target call event, that is, the conversation content of the target object. In some embodiments, the call content information is in the form of text information. It should be understood that the voice robot will conduct multiple rounds of guided conversations with the target object, and the call content information of the target object is also the reply content made by the target object to the voice robot. The emotion indication information represents the emotional characteristics of the target object in the target call event. In some embodiments, the emotion indication information is determined based on the conversation tone of the target object, for example, the volume of the conversation, the speed of the conversation, etc. In other embodiments, the emotion indication information is determined based on the emotional text of the target object, for example, a text containing target emotion words, such as a text containing satisfaction and happiness, or a text containing anger and annoyance.

[0095] In some embodiments, the process of the server determining the safety reference value of the historical call event based on the call content information and the emotion indication information of the target object includes any one of the following (302A) and (302B):

[0096] (302A) In some embodiments, the server uses a security identification model to determine the security reference value of the historical call event. The corresponding process is: the server inputs the call content information and emotion indication information of the target object into the security identification model, and predicts the security reference value of the historical call event of the target object through the security identification model to obtain the security reference value of the historical call event of the target object. In this way, the security reference value of the historical call event can be determined by using the security identification model, and the security reference value of the historical call event can be determined quickly, thereby improving the efficiency and accuracy of determining the security reference value.

[0097] The security identification model is used to predict the security reference value of the target object's historical call events. In some embodiments, the security identification model is a deep learning model, which is a model that gradually extracts higher-level features from the original input based on multiple processing layers containing complex structures or composed of multiple nonlinear transformations. For example, the security identification model can be a deep convolutional neural network.

[0098] Taking the security identification model as a deep convolutional neural network as an example, the security identification model includes an input layer, a convolution layer, a pooling layer, a fully connected layer and an output layer. Among them, the input layer is used to input the call content information and emotion indication information obtained by the server into the security identification model, and convert the input information into a digital matrix so that the security identification model can perform subsequent operations; the convolution layer is used to perform a convolution operation on the matrix generated by the input layer, and extract local features based on the result of the convolution operation. The security identification model may include one or more convolution layers; the pooling layer is used to quantize the feature extraction value obtained by the convolution layer to obtain a matrix with a smaller dimension, so as to further extract features. The security identification model may include one or more pooling layers; the fully connected layer is used to integrate the extracted local features into complete features through the weight matrix, and calculate the security reference value based on the complete feature; the output layer is used to obtain the security reference value output by the fully connected layer, and output the security reference value as the security reference value of the historical call event.

[0099] The security identification model adopted in the embodiments of the present disclosure is a trained model. In some embodiments, the security identification model is obtained by training based on the call content information, emotion indication information and security reference values ​​of multiple sample objects. In some embodiments, the server obtains the call content information, emotion indication information and security reference values ​​of the multiple sample objects in historical call events of multiple sample objects, and performs model training based on the call content information, emotion indication information and security reference values ​​of the multiple sample objects in historical call events of the multiple sample objects to obtain the security identification model. Specifically, the training process of the security identification model includes: in the first iteration process, the call content information and emotion indication information of the multiple sample objects are respectively input into the initial model to obtain the security reference value of the first iteration process; based on the security reference value of the first iteration process and the security reference value of the sample object, a loss function is determined, and based on the loss function, the model parameters in the initial model are adjusted; the model parameters adjusted in the first iteration are used as the model parameters of the second iteration, and the second iteration is performed again; the above iteration process is repeated multiple times, and in the Nth process, the model parameters adjusted in the N-1th iteration are used as new model parameters to perform model training until the training meets the target condition, and the model corresponding to the iteration process that meets the target condition is obtained as the security identification model. Wherein, N is a positive integer, and N is greater than 1. Optionally, the target condition satisfied by the training may be that the number of training iterations of the initial model reaches a target number, and the target number may be a preset number of training iterations; or, the target condition satisfied by the training may be that the loss value satisfies a target threshold condition, such as a loss value less than 0.00001. This is not limited in the embodiments of the present disclosure.

[0100] (302B) In some embodiments, the server determines the security reference value corresponding to the call content information of the target object based on the call content information of the target object and the first relationship data, and the first relationship data represents the correspondence between the call content information and the security reference value; the server determines the security reference value corresponding to the emotion indication information of the target object based on the emotion indication information of the target object and the second relationship data, and the second relationship data represents the correspondence between the emotion indication information and the security reference value; further, based on the security reference value corresponding to the call content information and the security reference value corresponding to the emotion indication information, the server determines the security reference value of the historical call events of the target object.

[0101] In some embodiments, the call content information of the target object is a security keyword included in the call content of the target object, and the security keyword represents a keyword related to security issues, such as "already remitted, deceived, property loss", etc. Accordingly, the first relationship data provides a corresponding relationship between the security keyword and the security reference value. In some embodiments, the emotion indication information is an emotion keyword used to characterize the emotional characteristics of the target object, and the emotion keyword represents a keyword related to emotion, such as "annoyance, negativity", etc. Accordingly, the second relationship data provides a corresponding relationship between the emotion keyword and the security reference value.

[0102] It should be noted that the first relationship data and the second relationship data are both relationship data associated with the server. In some embodiments, before implementing this solution, the server generates and stores the first relationship data based on a plurality of pre-set types of call content information and corresponding security reference values. In an optional embodiment, the server generates the first relationship data based on a plurality of pre-set security keywords and corresponding security reference values. In some embodiments, before implementing this solution, the server generates and stores the second relationship data based on a plurality of pre-set types of emotion indication information and corresponding security reference values. In an optional embodiment, the server generates the second relationship data based on a plurality of pre-set emotion keywords and corresponding security reference values.

[0103] In some embodiments, the server determines the security reference value of the historical call events of the target object based on the security reference value corresponding to the call content information and the security reference value corresponding to the emotion indication information. The process is as follows: based on the weight coefficient corresponding to the call content information and the weight coefficient corresponding to the emotion indication information, the security reference value corresponding to the call content information and the security reference value corresponding to the emotion indication information are weightedly summed to obtain the security reference value of the historical call events of the target object.

[0104] The above (302B) is a process of determining the security reference value of the historical call events of the target object by combining the call content information and emotion indication information of the target object. In other embodiments, the server can also determine the security reference value of the historical call events of the target object based on any one of the call content information and emotion indication information of the target object, for example, determining the security reference value of the historical call events of the target object based on the call content information of the target object, or determining the security reference value of the historical call events of the target object based on the emotion indication information of the target object.

[0105] 303. If the security reference value does not meet the security condition, send call information of the target call event to the management terminal, where the call information includes call content information of the first call audio.

[0106] Among them, the security condition is a pre-set fixed condition, and the security condition is used to determine the degree of possibility that the historical call event has a security problem. For example, satisfying the security condition indicates that the historical call event has a low degree of possibility of a security problem, and not satisfying the security condition indicates that the historical call event has a high degree of possibility of a security problem. The call content information of the first call audio includes the call content information of the target object and the call content information of the voice robot. It can be understood that the call content information of the first call audio is also the conversation content between the target object and the voice robot. Taking the call content information in the form of text information as an example, the call content information of the first call audio is also the text information of the first call audio.

[0107] In some embodiments, the safety condition is that the safety reference value is within the safety threshold interval. Accordingly, if the safety reference value is not within the safety threshold interval, the call information of the target call event is sent to the management terminal. For example, taking the safety reference value as an example of the level of the safety problem, the safety threshold interval can be a level interval [1, 3] formed by level 1 to level 3; or taking the safety reference value as an example of the probability of the safety problem, the safety threshold interval can be a probability interval [0, 30%] formed by probability 0 to probability 30%.

[0108] 304. The management terminal receives the call information of the target call event and displays the call information of the target call event in the call management interface.

[0109] The management terminal is a terminal operated by a manager. In some embodiments, the management terminal is provided with a visual interface of the communication management platform so that the manager can view and analyze.

[0110] For example, Figure 4 is a schematic diagram of a call management interface provided by an embodiment of the present disclosure, see Figure 4 ,exist Figure 4 In the illustrated call management interface, call information of the target call event is displayed. In some embodiments, the management terminal displays the call information of the target call event in the session area of ​​the call management interface. Figure 4 , the session area can be Figure 4 In some embodiments, the management terminal displays the call information of the target call event in the form of a dialogue in the conversation area of ​​the call management interface. Figure 4 On the left side of the conversation area of ​​the call management interface, the call content information of the target object in the target call event is displayed, and on the right side of the conversation area of ​​the call management interface, the call content information of the voice robot in the target call event is displayed.

[0111] The above steps 303 to 304 take the call information including the call content information of the first call audio as an example. In other embodiments, the call information also includes the call details information of the target call event. Then, the management terminal displays the call details information of the target call event in the call management interface, increasing the amount of information displayed in the call management interface, thereby providing managers with more abundant reference information for subsequent manual calls. For the corresponding content, see (304A) to (304D) below:

[0112] (304A) In some embodiments, the call information also includes the first call audio of the target call event. Accordingly, the management terminal displays the call content information of the first call audio in the call management interface, and also displays the first call audio and a trigger control of the first call audio.

[0113] For example, see Figure 4 ,exist Figure 4 In the call management interface shown, the first call audio can be Figure 4 The "audio" shown, the trigger control of the first call audio can be Figure 4 The “play control” shown. In this way, the amount of information displayed on the call management interface is increased, and the operation experience of managers is improved.

[0114] (304B) In some embodiments, the call information also includes basic information of the target call event, such as the target object's mobile phone number, call time, call duration, number of calls, and identity information used by the voice robot. Accordingly, the management terminal displays the call content information of the first call audio in the call management interface, and also displays the basic information of the target call event.

[0115] For example, see Figure 4 ,exist Figure 4In the call management interface shown, the target object's mobile phone number can be Figure 4 The "outbound call number: XXX" is shown, and the call time can be Figure 4 The displayed "call time: 2022-12-01 15:00:00" can be Figure 4 The displayed "call duration: 119" can be Figure 4 The “Number of interactions: 11” shown, the identity information used by the voice robot can be Figure 4 The "speech template: ×× District - Public Security, Procuratorate and Court" shown.

[0116] (304C) In some embodiments, the call information also includes a safety reference value of a historical call event of the target object. Accordingly, the management terminal displays the call content information of the first call audio in the call management interface, and also displays the safety reference value of the historical call event.

[0117] For example, see Figure 4 ,exist Figure 4 In the call management interface shown, the safety reference value of the target object's historical call events can be Figure 4 “Intelligent Assessment: High Risk” is shown.

[0118] (304D) In ​​some embodiments, the call information further includes the state information of the target object, and the state information indicates the state of the target object in the target call event, for example, being deceived, being difficult to persuade, being at high risk of being deceived, or not being deceived.

[0119] Accordingly, the contents of step 303 to step 304 can be replaced as follows: if the security reference value does not meet the security conditions, the server determines the status information of the target object based on the first call audio, and sends the call information carrying the status information to the management terminal. The management terminal then receives the call information carrying the status information, and displays the call content information of the first call audio in the call management interface, while also displaying the status information of the target object.

[0120] In some embodiments, the process by which the server determines the status information of the target object based on the first call audio is as follows: the server performs semantic recognition on the call content information of the first call audio to obtain the status information of the target object. In some embodiments, the server uses a first semantic recognition model to perform semantic recognition, and the corresponding process is as follows: the server inputs the call content information of the first call audio into the first semantic recognition model, performs semantic recognition on the call content information of the first call audio through the first semantic recognition model, and obtains the status information of the target object.

[0121] Semantic recognition refers to identifying the meaning of a word, a phrase or a text. The first semantic recognition model is used to judge the state information of the target object. Taking the state information including "deceived, difficult to persuade, high risk of being deceived" as an example, the first semantic recognition model is a three-classification model.

[0122] In some embodiments, the first semantic recognition model is obtained based on deep neural network training. Regarding the model structure and training process of the first semantic recognition model, refer to the model structure and training process of the security recognition model in step 302, which will not be repeated here.

[0123] For example, see Figure 4 ,exist Figure 4 In the call management interface shown, the status information of the target object can be Figure 4 "Risk Type: Defrauded" is shown.

[0124] With respect to the detailed information of the target call event shown in (304A) to (304D) above, in some embodiments, the management terminal displays the detailed information of the target call event in the call details area of ​​the call management interface. Figure 4 , the call details area can be Figure 4 "Area 402" is shown.

[0125] 305. The server controls the voice robot to play target audio data, where the target audio data is used to prompt that the manual call mode is about to be used for the call.

[0126] In some embodiments, the server controls the voice robot to use anthropomorphic sound effects to play the target audio data to prompt the target object that the call will be made in the manual call mode. For example, the content of the target audio data includes "Please wait, I will let our on-duty person come to check with you in detail." In this way, by playing the target audio data, the target object can know the progress of the call in time, thereby improving the user's call experience.

[0127] In the above steps 303 to 305, when the safety reference value does not meet the safety condition, the solution is described by first sending the call information of the target call event to the management terminal, and then controlling the voice robot to play the target voice data as an example, and in other embodiments, when the safety reference value does not meet the safety condition, the server first controls the voice robot to play the target voice data, and then sends the call information of the target call event to the management terminal, or the server sends the call information of the target call event to the management terminal while controlling the voice robot to play the target voice data. The embodiment of the present disclosure does not limit the execution order of the above steps 303 to 305.

[0128] 306. The server switches to manual call mode.

[0129] In some embodiments, the server sends a call request to the management terminal. In response to the management terminal accepting the call request, the call is successfully switched to the manual call mode. At this time, the manager can continue to talk to the target object based on the call content of the voice robot. In this way, in the process of talking based on the voice robot, a method of automatically switching to the manual call mode is realized, which improves the efficiency of human-computer interaction.

[0130] In some embodiments, the server also sends a call prompt message to the management terminal, where the call prompt message is used to prompt that there is a call event to be connected. Accordingly, the management terminal receives the call prompt message and displays the call prompt message in the call management interface.

[0131] There can be other implementations for the management terminal mentioned in the above embodiments. In some embodiments, the communication management platform is associated with at least one management terminal. After executing step 302, if the security reference value does not meet the security condition, the server selects a management terminal from at least one associated management terminal, and then executes the subsequent call processing flow based on the selected management terminal. In an optional embodiment, the server selects a management terminal that is in an idle state from at least one associated management terminal. In this way, by selecting a management terminal that is in an idle state for subsequent calls, the call success rate is improved and the effect of subsequent manual dissuasion is ensured.

[0132] 307. The server extracts call content information corresponding to the manual call mode from the second call audio of the manual call mode.

[0133] The second call audio refers to the call audio when the manual call mode is used. The call content information corresponding to the manual call mode refers to the call content fed back by the manager in the target call event, that is, the conversation content of the manager. Taking the call content information in the form of text information as an example, the call content information corresponding to the manual call mode is also the text information corresponding to the manual call mode.

[0134] In some embodiments, the server extracts the call audio of the manager from the second call audio based on the audio features of the manager, performs audio recognition on the extracted call audio, and obtains the call content information corresponding to the manager, that is, extracts the call content information corresponding to the manual call mode. Among them, the audio feature can be a feature used to characterize the timbre. In some embodiments, the server is provided with an audio-to-text converter, through which the extracted call audio is subjected to audio recognition, and the call content information of the call audio can be obtained. In other embodiments, the server performs semantic recognition on the call content information of the second call audio to obtain the call content information corresponding to the manual call mode. In the above embodiments, two methods of extracting the call content information corresponding to the manual call mode are provided, which improves the flexibility of extracting the call content information corresponding to the manual call mode.

[0135] 308. The server determines a quality reference value of the call content information corresponding to the manual call mode, where the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode.

[0136] In some embodiments, the quality reference value is in the form of a quality score. In some embodiments, the process of the server determining the quality reference value includes at least one of the following:

[0137] (308A) In some embodiments, the server determines a quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event, where the call effect information indicates the effect of the safety prompt of the target call event.

[0138] In some embodiments, the call effect information is used to indicate whether the target call event is successfully dissuaded, or the call effect information is used to indicate the level of the dissuasion effect of the target call event, for example, excellent, good, or poor.

[0139] In some embodiments, the server determines the quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event and the third relationship data, and the third relationship data represents the correspondence between the call effect information and the quality reference value. It should be noted that the third relationship data is the relationship data associated with the server. In some embodiments, before implementing this solution, the server generates and stores the third relationship data based on multiple types of pre-set call effect information and corresponding quality reference values.

[0140] In some embodiments, the process of the server acquiring the call effect information is as follows: the server performs semantic recognition on the call content information of the second call audio of the manual call mode to obtain the call effect information of the second call audio.

[0141] In some embodiments, the server uses a second semantic recognition model to perform semantic recognition, and the corresponding process is: the server inputs the call content information of the second call audio into the second semantic recognition model, performs semantic recognition on the call content information of the second call audio through the second semantic recognition model, and obtains the call effect information of the second call audio.

[0142] The second semantic recognition model is used to judge the call effect (i.e., dissuasion effect) of the second call audio. Taking the case where the call effect information is used to indicate whether the dissuasion of this call is successful as an example, the second semantic recognition model is a binary classification model; taking the case where the call effect information is used to indicate the level of the dissuasion effect of this call as an example, the second semantic recognition model is a multi-classification model.

[0143] In some embodiments, the second semantic recognition model is obtained based on deep neural network training. Regarding the model structure and training process of the second semantic recognition model, refer to the model structure and training process of the security recognition model in step 302, which will not be repeated here.

[0144] (308B) In some embodiments, the server determines a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event.

[0145] In some embodiments, the server determines the quality reference value of the call content information corresponding to the artificial call mode based on the first matching degree and fourth relationship data between the call content information corresponding to the artificial call mode and the target call event, and the fourth relationship data represents the corresponding relationship between the first matching degree and the quality reference value.

[0146] Among them, the first matching degree represents the degree of correlation between the call content information corresponding to the manual call mode and the target call event. For example, the higher the degree of correlation between the call content information corresponding to the manual call mode and the target call event, the higher the quality reference value of the call content information corresponding to the manual call mode. It should be noted that the fourth relationship data is the relationship data associated with the server. In some embodiments, before implementing this solution, the server generates and stores the fourth relationship data based on a plurality of pre-set first matching degrees and corresponding quality reference values.

[0147] In some embodiments, the first matching degree is the matching degree between the call content information corresponding to the manual call mode and the status information of the target call event. Accordingly, the server determines the quality reference value of the call content information corresponding to the manual call mode based on the matching degree between the call content information corresponding to the manual call mode and the status information of the target call event.

[0148] (308C) In some embodiments, the server determines a quality reference value of the call content information corresponding to the manual call mode based on a second matching degree between the call content information corresponding to the manual call mode and the historical call event.

[0149] In some embodiments, the server determines the quality reference value of the call content information corresponding to the manual call mode based on the second matching degree and fifth relationship data between the call content information corresponding to the manual call mode and the historical call event, and the fifth relationship data represents the corresponding relationship between the second matching degree and the quality reference value.

[0150] Among them, the second matching degree indicates the degree of correlation between the call content information corresponding to the manual call mode and the historical call event. For example, the higher the degree of correlation between the call content information corresponding to the manual call mode and the historical call event, the higher the quality reference value of the call content information corresponding to the manual call mode. It should be noted that the fifth relationship data is the relationship data associated with the server. In some embodiments, before implementing this solution, the server generates and stores the fifth relationship data based on a plurality of pre-set second matching degrees and corresponding quality reference values.

[0151] In some embodiments, the second matching degree is the matching degree between the call content information corresponding to the manual call mode and the event type of the historical call event, and the event type indicates the content type of the event content of the historical call event. Taking telecommunications network fraud as an example, the event type is also the fraud type (or fraud method). Accordingly, the server determines the quality reference value of the call content information corresponding to the manual call mode based on the matching degree between the call content information corresponding to the manual call mode and the event type of the historical call event.

[0152] (308D) In ​​some embodiments, the server determines a quality reference value of the call content information corresponding to the artificial call mode based on the rarity of the call content information corresponding to the artificial call mode, where the rarity indicates the rarity of the corresponding call content information.

[0153] In some embodiments, the server determines the quality reference value of the call content information corresponding to the artificial call mode based on the rarity of the call content information corresponding to the artificial call mode and the sixth relationship data, and the sixth relationship data represents the corresponding relationship between the rarity and the quality reference value. It should be noted that the sixth relationship data is the relationship data associated with the server. In some embodiments, before implementing this solution, the server generates and stores the sixth relationship data based on a plurality of pre-set rarities and corresponding quality reference values.

[0154] For example, the higher the rarity of the call content information corresponding to the manual call mode, the higher the quality reference value of the call content information corresponding to the manual call mode. In some embodiments, the process of the server determining the rarity is as follows: the server determines the content occurrence frequency of the call content information corresponding to the manual call mode based on a plurality of content information for security prompts included in the content information library, and determines the determined content occurrence frequency as the rarity, where the content occurrence frequency indicates the frequency of the call content information corresponding to the manual call mode appearing in the content information library.

[0155] In the above embodiments, multiple ways of determining the quality reference value are shown, which improves the flexibility of determining the quality reference value.

[0156] 309. If the quality reference value meets the content quality condition, the server updates the content information library based on the call content information corresponding to the manual call mode, and the content information library includes a plurality of content information for security prompts.

[0157] The content quality condition is a pre-set fixed condition, and the content quality condition is used to determine whether the content quality of the call content information corresponding to the manual call mode meets the quality requirement. In some embodiments, the content quality condition is that the quality reference value reaches the quality score threshold.

[0158] In some embodiments, if the quality reference value meets the content quality condition, the server adds the call content information corresponding to the manual call mode in the content information library, so as to subsequently perform call dissuasion based on the call content information corresponding to the manual call mode.

[0159] It should be noted that the above steps 307 to 309 are optional steps. In some embodiments, after the server executes step 306, there is no need to execute steps 307 to 309.

[0160] In the process of determining the quality reference value of the call content information corresponding to the manual call mode in the above embodiment, the call effect information of the target call event is involved, see (308A). In some embodiments, after executing step 306, the server executes the step of determining the call effect information of the target call event. Further optionally, after determining the call effect information of the target call event, the server also adds a call effect label to the target call event based on the call effect information of the target call event. For example, the call effect label can be "poor dissuasion effect". In this way, by adding a call effect label to the target call event, the management personnel can timely know the call effect of this call event. In some embodiments, after executing step 308, the server executes the step of determining the call effect information of the target call event. In an optional embodiment, for multiple call events, the server determines the call effect information of the corresponding call events according to the quality reference values ​​of the call content information corresponding to the manual call mode in the multiple call events from low to high. In this way, the call effect information is determined preferentially for the call event with a lower quality reference value, so as to timely know the call effect of the corresponding call event, so as to perform subsequent processing for the call event with poor call effect.

[0161] The technical solution provided by the embodiment of the present disclosure first adopts the call mode of a voice robot to talk to the target object in a target call event triggered by the historical call event of the target object. In this way, when facing a large number of telecommunications network frauds, the call mode of the voice robot is preferentially adopted, which can effectively reduce the call pressure and improve the call efficiency. During the call, based on the call audio between the voice robot and the target object, the safety reference value of the target object in the historical call event is determined. When the safety reference value does not meet the safety conditions, the manual call mode is adopted to talk to the target object to ensure the call efficiency. In this way, a human-machine combined call dissuasion method is provided, which can improve the dissuasion efficiency while ensuring the dissuasion effect.

[0162] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present disclosure, and will not be described in detail here.

[0163] Figure 5 is a structural diagram of a call processing device provided by an embodiment of the present disclosure, see Figure 5 , the device comprises:

[0164] A call module 501 is used to control the voice robot to talk to the target object in a target call event triggered by a historical call event of the target object, wherein the target call event is used to provide a safety prompt to the target object for the historical call event;

[0165] A determination module 502, configured to determine a safety reference value of a historical call event of the target object based on a first call audio of the call between the voice robot and the target object;

[0166] The switching module 503 is used to switch to the manual call mode if the safety reference value does not meet the safety condition.

[0167] The technical solution provided by the embodiment of the present disclosure first adopts the call mode of a voice robot to talk to the target object in a target call event triggered by the historical call event of the target object. In this way, when facing a large number of telecommunications network frauds, the call mode of the voice robot is preferentially adopted, which can effectively reduce the call pressure and improve the call efficiency. During the call, based on the call audio between the voice robot and the target object, the safety reference value of the target object in the historical call event is determined. When the safety reference value does not meet the safety conditions, the manual call mode is adopted to talk to the target object to ensure the call efficiency. In this way, a human-machine combined call dissuasion method is provided, which can improve the dissuasion efficiency while ensuring the dissuasion effect.

[0168] In some embodiments, the determination module 502 includes:

[0169] An extraction submodule, configured to extract call content information and / or emotion indication information of the target object from the first call audio, wherein the emotion indication information represents an emotional feature of the target object in the target call event;

[0170] The determination submodule is used to determine the safety reference value of the historical call events of the target object based on the call content information and / or emotion indication information of the target object.

[0171] In some embodiments, a submodule is determined to:

[0172] The call content information and emotion indication information of the target object are input into a security recognition model, and the security reference value of the historical call events of the target object is predicted by the security recognition model to obtain the security reference value of the historical call events of the target object. The security recognition model is trained based on the call content information, emotion indication information and security reference value of multiple sample objects.

[0173] In some embodiments, the apparatus further comprises:

[0174] The sending module is used to send the call information of the target call event to the management terminal, and the call information includes the call content information of the first call audio.

[0175] In some embodiments, the call information further includes state information of the target object, the state information indicating the state of the target object in the target call event;

[0176] The determination module 502 is further configured to:

[0177] If the safety reference value does not meet the safety condition, the status information of the target object is determined based on the first call audio.

[0178] In some embodiments, the apparatus further comprises:

[0179] The control module is used to control the voice robot to play target audio data, and the target audio data is used to prompt that the manual call mode is about to be used for the call.

[0180] In some embodiments, the apparatus further comprises:

[0181] The identification module is used to perform semantic recognition on the call content information of the second call audio of the manual call mode to obtain the call effect information of the second call audio, and the call effect information indicates the effect of the security prompt of the target call event.

[0182] In some embodiments, the device further includes an extraction module for extracting call content information corresponding to the manual call mode from the second call audio of the manual call mode;

[0183] The determination module 502 is further configured to determine a quality reference value of the call content information corresponding to the manual call mode, where the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode.

[0184] In some embodiments, the determination module 502 is further configured to:

[0185] Determine a quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event, wherein the call effect information indicates the effect of the safety prompt of the target call event;

[0186] Determining a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event;

[0187] Determining a quality reference value of the call content information corresponding to the manual call mode based on a second matching degree between the call content information corresponding to the manual call mode and the historical call event;

[0188] Based on the rarity of the call content information corresponding to the manual call mode, a quality reference value of the call content information corresponding to the manual call mode is determined, where the rarity indicates the rarity of the corresponding call content information.

[0189] In some embodiments, the apparatus further comprises:

[0190] The updating module is used to update the content information library based on the call content information corresponding to the manual call mode if the quality reference value meets the content quality condition. The content information library includes multiple content information for security prompts.

[0191] It should be noted that: the call processing device provided in the above embodiment only uses the division of the above functional modules as an example when performing call processing. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the electronic device is divided into different functional modules to complete all or part of the functions described above. In addition, the call processing device provided in the above embodiment and the call processing method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0192] The electronic device in the embodiments of the present disclosure may be provided as a terminal, and specifically may be the management terminal mentioned in the above embodiments. Figure 6 600 is a schematic diagram of the structure of a terminal 600 provided in an embodiment of the present disclosure. The terminal 600 may be: a smart device, a tablet computer, a laptop computer or a desktop computer. The terminal 600 may also be called a user device, a portable terminal, a laptop terminal, a desktop terminal or other names.

[0193] Typically, the terminal 600 includes a processor 601 and a memory 602 .

[0194] The processor 601 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 601 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 601 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 601 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 601 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0195] The memory 602 may include one or more computer-readable storage media, which may be non-transitory. The memory 602 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 602 is used to store at least one program code, which is used to be executed by the processor 601 to implement the call processing method provided by the management terminal in the method embodiment of the present disclosure.

[0196] In some embodiments, the terminal 600 may further optionally include: a peripheral device interface 603 and at least one peripheral device. The processor 601, the memory 602 and the peripheral device interface 603 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 603 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 604, a display screen 605, a camera assembly 606, an audio circuit 607, a positioning assembly 608 and a power supply 609.

[0197] The peripheral device interface 603 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 601 and the memory 602. In some embodiments, the processor 601, the memory 602, and the peripheral device interface 603 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 601, the memory 602, and the peripheral device interface 603 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0198] The radio frequency circuit 604 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 604 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 604 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 604 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and the like. The radio frequency circuit 604 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: a metropolitan area network, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 604 may also include circuits related to NFC (Near Field Communication), which is not limited in the present disclosure.

[0199] The display screen 605 is used to display the UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 605 is a touch display screen, the display screen 605 also has the ability to collect touch signals on the surface or above the surface of the display screen 605. The touch signal can be input to the processor 601 as a control signal for processing. At this time, the display screen 605 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 605 can be one, set on the front panel of the terminal 600; in other embodiments, the display screen 605 can be at least two, respectively set on different surfaces of the terminal 500 or in a folding design; in other embodiments, the display screen 605 can be a flexible display screen, set on the curved surface or folding surface of the terminal 600. Even, the display screen 605 can also be set to a non-rectangular irregular shape, that is, a special-shaped screen. The display screen 605 can be made of materials such as LCD (Liquid Crystal Display), OLED (Organic Light-Emitting Diode, organic light-emitting diode).

[0200] The camera assembly 606 is used to capture images or videos. Optionally, the camera assembly 606 includes a front camera and a rear camera. Typically, the front camera is arranged on the front panel of the terminal, and the rear camera is arranged on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, a wide-angle camera, and a telephoto camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, the fusion of the main camera and the wide-angle camera to realize the panoramic shooting and VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera assembly 606 may also include a flash. The flash can be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.

[0201] The audio circuit 607 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 601 for processing, or input them into the radio frequency circuit 604 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 600. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 601 or the radio frequency circuit 604 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 607 may also include a headphone jack.

[0202] Positioning component 608 is used to locate the current geographical location of terminal 600 to implement navigation or LBS (Location Based Service). Positioning component 608 can be a positioning component based on the US GPS (Global Positioning System), China's Beidou system, Russia's Grenas system or the European Union's Galileo system.

[0203] The power supply 609 is used to power various components in the terminal 600. The power supply 609 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 609 includes a rechargeable battery, the rechargeable battery can support wired charging or wireless charging. The rechargeable battery can also be used to support fast charging technology.

[0204] In some embodiments, the terminal 600 further includes one or more sensors 610 , including but not limited to: an acceleration sensor 611 , a gyroscope sensor 612 , a pressure sensor 613 , a fingerprint sensor 614 , an optical sensor 615 , and a proximity sensor 616 .

[0205] The acceleration sensor 611 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the terminal 600. For example, the acceleration sensor 611 can be used to detect the components of gravity acceleration on the three coordinate axes. The processor 601 can control the display screen 605 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 611. The acceleration sensor 611 can also be used for collecting game or user motion data.

[0206] The gyro sensor 612 can detect the body direction and rotation angle of the terminal 600, and the gyro sensor 612 can cooperate with the acceleration sensor 611 to collect the user's 3D actions on the terminal 600. The processor 601 can implement the following functions based on the data collected by the gyro sensor 612: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.

[0207] The pressure sensor 613 can be set on the side frame of the terminal 600 and / or the lower layer of the display screen 605. When the pressure sensor 613 is set on the side frame of the terminal 600, it can detect the user's holding signal of the terminal 600, and the processor 601 performs left and right hand recognition or shortcut operation according to the holding signal collected by the pressure sensor 613. When the pressure sensor 613 is set on the lower layer of the display screen 605, the processor 601 controls the operability controls on the UI interface according to the user's pressure operation on the display screen 605. The operability controls include at least one of a button control, a scroll bar control, an icon control, and a menu control.

[0208] The fingerprint sensor 614 is used to collect the user's fingerprint, and the processor 601 identifies the user's identity based on the fingerprint collected by the fingerprint sensor 614, or the fingerprint sensor 614 identifies the user's identity based on the collected fingerprint. When the user's identity is identified as a trusted identity, the processor 601 authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings. The fingerprint sensor 614 can be set on the front, back, or side of the terminal 600. When a physical button or a manufacturer logo is set on the terminal 600, the fingerprint sensor 614 can be integrated with the physical button or the manufacturer logo.

[0209] The optical sensor 615 is used to collect the ambient light intensity. In one embodiment, the processor 601 can control the display brightness of the display screen 605 according to the ambient light intensity collected by the optical sensor 615. Specifically, when the ambient light intensity is high, the display brightness of the display screen 605 is increased; when the ambient light intensity is low, the display brightness of the display screen 605 is reduced. In another embodiment, the processor 601 can also dynamically adjust the shooting parameters of the camera assembly 606 according to the ambient light intensity collected by the optical sensor 615.

[0210] The proximity sensor 616, also called a distance sensor, is usually arranged on the front panel of the terminal 600. The proximity sensor 616 is used to collect the distance between the user and the front of the terminal 600. In one embodiment, when the proximity sensor 616 detects that the distance between the user and the front of the terminal 600 is gradually decreasing, the processor 601 controls the display screen 605 to switch from the screen-on state to the screen-off state; when the proximity sensor 616 detects that the distance between the user and the front of the terminal 600 is gradually increasing, the processor 601 controls the display screen 605 to switch from the screen-off state to the screen-on state.

[0211] Those skilled in the art will understand that Figure 6 The structure shown in the figure does not constitute a limitation on the terminal 600, and the terminal 600 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0212] The electronic device in the embodiment of the present disclosure can be provided as a server. Figure 7 It is a structural diagram of a server provided by an embodiment of the present disclosure. The server 700 may have relatively large differences due to different configurations or performances, and may include one or more processors (Central Processing Units, CPU) 701 and one or more memories 702, wherein the one or more memories 702 store at least one program code, and the at least one program code is loaded and executed by the one or more processors 701 to implement the call processing method executed by the server in the above-mentioned various method embodiments. Of course, the server 700 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The server 700 may also include other components for implementing device functions, which will not be described in detail here.

[0213] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a program code, and the program code can be executed by a processor to complete the call processing method in the above embodiment. For example, the computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, etc.

[0214] In an exemplary embodiment, a computer program product is also provided, which includes a computer program code, which is stored in a computer-readable storage medium. A processor of an electronic device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, so that the electronic device performs the above-mentioned call processing method.

[0215] In some embodiments, the computer program involved in the embodiments of the present disclosure may be deployed and executed on one electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed at multiple locations and interconnected by a communication network. Multiple electronic devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.

[0216] A person skilled in the art will appreciate that all or part of the steps for implementing the above embodiments may be accomplished by hardware or by hardware associated with a program code, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0217] The above are only optional embodiments of the present disclosure and are not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A call processing method, characterized in that: The method comprises: In a target call event triggered by a historical call event of the target object, controlling the voice robot to talk to the target object, wherein the target call event is used to provide a safety prompt to the target object for the historical call event; Extracting call content information and emotion indication information of the target object from a first call audio of a call between the voice robot and the target object, wherein the emotion indication information represents an emotional feature of the target object in the target call event, and the emotion indication information is determined based on the conversation tone or emotional text of the target object; Determining a security reference value corresponding to the call content information of the target object based on the call content information of the target object and first relationship data, wherein the first relationship data represents a corresponding relationship between the call content information and the security reference value; Determining a safety reference value corresponding to the emotion indication information of the target object based on the emotion indication information of the target object and second relationship data, wherein the second relationship data represents a corresponding relationship between the emotion indication information and the safety reference value; Based on the weight coefficient corresponding to the call content information and the weight coefficient corresponding to the emotion indication information, a weighted sum is performed on the safety reference value corresponding to the call content information and the safety reference value corresponding to the emotion indication information to obtain the safety reference value of the historical call event of the target object; If the safety reference value does not meet the safety condition, switch to manual call mode; Extracting call content information corresponding to the manual call mode from the second call audio of the manual call mode; Determine a quality reference value of the call content information corresponding to the manual call mode, wherein the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode; Wherein, determining the quality reference value of the call content information corresponding to the manual call mode includes at least one of the following: Determine a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event, wherein the first matching degree is a matching degree between the call content information corresponding to the manual call mode and the state information of the target call event; Based on the second matching degree between the call content information corresponding to the manual call mode and the historical call event, the quality reference value of the call content information corresponding to the manual call mode is determined, and the second matching degree is the matching degree between the call content information corresponding to the manual call mode and the event type of the historical call event.

2. The method according to claim 1, characterized in that After determining that the safety reference value does not meet the safety condition and before switching to the manual call mode, the method further includes: Send call information of the target call event to a management terminal, where the call information includes call content information of the first call audio.

3. The method according to claim 2, characterized in that The call information further includes state information of the target object, wherein the state information indicates the state of the target object in the target call event; The method further comprises: If the safety reference value does not meet the safety condition, the status information of the target object is determined based on the first call audio.

4. The method according to claim 1, characterized in that After determining that the safety reference value does not meet the safety condition and before switching to the manual call mode, the method further includes: The voice robot is controlled to play target audio data, where the target audio data is used to prompt that the manual call mode is about to be used for a call.

5. The method according to claim 1, characterized in that After switching to the manual call mode, the method further includes: Semantic recognition is performed on the call content information of the second call audio of the manual call mode to obtain call effect information of the second call audio, where the call effect information represents the effect of the security prompt of the target call event.

6. The method according to claim 1, characterized in that The step of determining the quality reference value of the call content information corresponding to the manual call mode further includes at least one of the following: Determine a quality reference value of the call content information corresponding to the manual call mode based on the call effect information of the target call event, wherein the call effect information indicates the effect of the safety prompt of the target call event; Based on the rarity of the call content information corresponding to the manual call mode, a quality reference value of the call content information corresponding to the manual call mode is determined, wherein the rarity indicates the rarity of the corresponding call content information.

7. The method according to claim 1, characterized in that After determining the quality reference value of the call content information corresponding to the manual call mode, the method further includes: If the quality reference value meets the content quality condition, the content information library is updated based on the call content information corresponding to the manual call mode, and the content information library includes a plurality of content information for security prompts.

8. A call processing device, characterized in that: The device comprises: A call module, used to control the voice robot to talk to the target object in a target call event triggered by a historical call event of the target object, wherein the target call event is used to provide a safety prompt to the target object for the historical call event; A determination module, used to extract call content information and emotion indication information of the target object from the first call audio of the call between the voice robot and the target object, the emotion indication information representing the emotional characteristics of the target object in the target call event, and the emotion indication information is determined based on the conversation tone or emotional text of the target object; determine a safety reference value corresponding to the call content information of the target object based on the call content information of the target object and first relationship data, the first relationship data representing the correspondence between the call content information and the safety reference value; determine a safety reference value corresponding to the emotion indication information of the target object based on the emotion indication information of the target object and second relationship data, the second relationship data representing the correspondence between the emotion indication information and the safety reference value; based on the weight coefficient corresponding to the call content information and the weight coefficient corresponding to the emotion indication information, perform weighted summation of the safety reference value corresponding to the call content information and the safety reference value corresponding to the emotion indication information to obtain the safety reference value of the historical call event of the target object; A switching module, used for switching to a manual call mode if the safety reference value does not meet the safety condition; An extraction module, used to extract call content information corresponding to the manual call mode from the second call audio of the manual call mode; The determination module is further used to determine a quality reference value of the call content information corresponding to the manual call mode, wherein the quality reference value is used to indicate the content quality of the call content information corresponding to the manual call mode; The determining module is used to perform at least one of the following: Determine a quality reference value of the call content information corresponding to the manual call mode based on a first matching degree between the call content information corresponding to the manual call mode and the target call event, wherein the first matching degree is a matching degree between the call content information corresponding to the manual call mode and the state information of the target call event; Based on the second matching degree between the call content information corresponding to the manual call mode and the historical call event, the quality reference value of the call content information corresponding to the manual call mode is determined, and the second matching degree is the matching degree between the call content information corresponding to the manual call mode and the event type of the historical call event.

9. An electronic device, characterized in that: The electronic device includes one or more processors and one or more memories, wherein the one or more memories store at least one program code, and the program code is loaded and executed by the one or more processors to implement the call processing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one program code, and the program code is loaded and executed by the processor to implement the call processing method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The computer program product includes a computer program code, which is stored in a computer-readable storage medium. A processor of an electronic device reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code, so that the electronic device executes the call processing method as described in any one of claims 1 to 7.

Images