Virtual call method, system and device and medium

Abstract

The embodiment of the invention provides a virtual call method and system, and the method comprises the steps: starting a virtual call in response to a triggering condition; the triggering condition comprises a triggering action and / or an environment state of the passenger; the virtual call comprises: generating a virtual active call content; virtual call response content is generated based on the passenger's response to the virtual active call content and / or the environmental state.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a virtual call method and system. Background Art

[0002] While riding in a car, users may encounter stressful situations and want to seek help from others, but there is a high probability that they will have difficulty identifying the right person to ask for help or will not be able to contact the other party in time, which will lead to increasing anxiety.

[0003] Therefore, it is desirable to provide a virtual call method to help users escape from tense situations or to warn people with bad intentions, so as to prevent users from falling into unfavorable situations. Summary of the invention

[0004] One of the embodiments of this specification provides a virtual call method, which is executed by a mobile terminal. The method includes: starting a virtual call in response to a trigger condition; the trigger condition includes a passenger's trigger action and / or an environmental state; the virtual call includes: generating a virtual active call content; based on the passenger's response to the virtual active call content and / or the environmental state, generating a virtual call response content.

[0005] In some embodiments, a virtual incoming call reminder may be generated based on the trigger condition.

[0006] In some embodiments, when the service platform conditions meet the requirements, the virtual call response content generated by the large language model can be obtained from the service platform.

[0007] In some embodiments, the text of the conversation can be generated based on pre-stored text fragments and conversation information through a lightweight model in the mobile terminal; and a virtual call voice can be generated based on the text of the conversation.

[0008] In some embodiments, the method may also include: in response to the passenger's response to the virtual proactive call content satisfying a first alarm condition, starting the recording or video recording software of the mobile terminal; based on the recording or video recording content recorded by the recording or video recording software, determining whether the passenger's response to the virtual proactive call content satisfies a second alarm condition; in response to the passenger's response to the virtual proactive call content satisfying the second alarm condition, issuing an alarm signal.

[0009] In some embodiments, based on the communication signal status and / or positioning information of the mobile terminal meeting preset conditions, the virtual call and / or simulated call quality change can be suspended to improve the authenticity of the virtual call.

[0010] One of the embodiments of the present specification provides a virtual call system, the system comprising: a detection module for detecting trigger conditions for a virtual call, the trigger conditions comprising a passenger's triggering action and / or environmental conditions; a virtual call generation module for generating virtual active call content based on the trigger conditions and generating virtual call response content based on the passenger's response to the virtual active call content and / or the environmental conditions.

[0011] In some embodiments, the virtual call generation module further includes a virtual incoming call reminder module, and the virtual call generation module is further used to: generate a virtual incoming call reminder based on the trigger condition.

[0012] In some embodiments, the system further includes: a service platform configured to generate the virtual call response content through a large language model when platform conditions meet the requirements.

[0013] In some embodiments, the virtual call generation module is further used to: generate the text of the conversation based on pre-stored text fragments and conversation information through a lightweight model in the mobile terminal; and generate a virtual call voice based on the text of the conversation.

[0014] In some embodiments, the system further includes a judgment module. The judgment module is used to: in response to the passenger's response to the virtual active call content meeting the first alarm condition, start the recording or video recording software of the mobile terminal; based on the recording or video recording content recorded by the recording or video recording software, judge whether the passenger's response to the virtual active call content meets the second alarm condition; in response to the passenger's response to the virtual active call content meeting the second alarm condition, send an alarm signal.

[0015] In some embodiments, the system further includes a call quality detection module, and the call quality detection module includes: pausing the virtual call and / or simulating call quality changes based on the communication signal status and / or positioning information of the mobile terminal meeting preset conditions.

[0016] One of the embodiments of the present specification provides a virtual call device, which includes at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least part of the computer instructions to implement the virtual call method as described above.

[0017] One or more embodiments of the present specification provide a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the virtual call method as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0019] Figure 1 is a schematic diagram of an application scenario of a virtual call system according to some embodiments of this specification;

[0020] Figure 2 is an exemplary schematic diagram of a virtual call system according to some embodiments of the present specification;

[0021] Figure 3 is an exemplary flow chart of a virtual call method according to some embodiments of this specification;

[0022] Figure 4 is an exemplary schematic diagram of a method for generating call response content according to some embodiments of this specification;

[0023] Figure 5 is an exemplary schematic diagram of a lightweight model according to some embodiments of the present specification;

[0024] Figure 6 is an exemplary schematic diagram of determining whether to issue an alarm signal according to some embodiments of this specification;

[0025] Figure 7 is an exemplary schematic diagram of guiding or prompting a user to make a virtual call according to some embodiments of this specification;

[0026] Figure 8 It is an exemplary schematic diagram of simulated call quality changes according to some embodiments of this specification. DETAILED DESCRIPTION

[0027] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0028] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0029] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0030] Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may also be added to these processes, or one or more operations may be removed from these processes.

[0031] Figure 1 1 is a schematic diagram of an application scenario of a virtual call system according to some embodiments of the present specification. In the present specification, the application scenario 100 of the virtual call system is referred to as the application scenario 100 hereinafter.

[0032] like Figure 1 As shown, in some embodiments, the application scenario 100 may include a processor 110 , a network 120 , a terminal 130 , and a storage device 140 .

[0033] The processor 110 may acquire and / or process various information and data. In some embodiments, the processor 110 may execute the virtual call method as described in some embodiments of this specification. In some embodiments, the processor 110 may start a virtual call in response to a trigger condition. In some embodiments, the processor 110 may generate virtual active call content, and generate virtual call response content based on the passenger's response to the virtual active call content and / or the environmental state.

[0034] In some embodiments, the processor 110 may exchange information and data with other components (e.g., the terminal 130, the storage device 140) in the application scenario 100 through the network 120. For example, a pre-stored text segment, time information, and location information, etc., for generating text of a virtual call conversation are obtained from the storage device 140. In some embodiments, the processor 110 may include multiple devices, among which the devices for obtaining the environment state, generating the virtual active call content, generating the virtual call response content, etc. may be the same device or different devices.

[0035] In some embodiments, processor 110 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, processor 110 may be implemented on a cloud platform. As an example only, a cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-layer cloud, etc., or any combination thereof.

[0036] The network 120 can connect the components of the system and / or connect the system with external resources. The network 120 enables communication between the components and with other parts outside the system to facilitate the exchange of data and / or information. In some embodiments, one or more components in the application scenario 100 (e.g., the processor 110, the terminal 130, the storage device 140) can send data and / or information to other components via the network 120. In some embodiments, the network 120 can be any one or more of a wired network or a wireless network.

[0037] The terminal 130 can be used to input / output information and / or data. In some embodiments, the terminal 130 can obtain the user's trigger action and the user's environmental state. In some embodiments, the terminal 130 can display to the user a virtual call response content generated by the processor 110. In some embodiments, the terminal 130 can be a mobile device 130-1, a tablet computer 130-2, a laptop computer 130-3, etc., and any combination of other devices with input and / or output functions.

[0038] The storage device 140 may be used to store data and / or instructions. In some embodiments, the storage device 140 may store data and / or information during the processing of the processor 110, such as time information, location information, etc. The storage device 140 may include one or more storage components, each of which may be an independent device or part of other components (e.g., the processor 110, the terminal 130) in the application scenario 100. The storage device may be local or implemented through the cloud.

[0039] It should be noted that application scenario 100 is provided only for the purpose of illustration and is not intended to limit the scope of this specification. For those of ordinary skill in the art, various modifications or changes can be made according to the description of this specification. For example, application scenario 100 can also include a database. For another example, application scenario 100 can be implemented on other devices to achieve similar or different functions. However, changes and modifications will not deviate from the scope of this specification.

[0040] Figure 2 It is an exemplary schematic diagram of a virtual call system according to some embodiments of the present specification.

[0041] like Figure 2 As shown, in some embodiments, the virtual call system 200 may include a detection module 210 and a virtual call generation module 220. In some embodiments, each module in the virtual call system 200 may be implemented by the processor 110.

[0042] In some embodiments, the detection module 210 may be used to detect a triggering condition for a virtual call.

[0043] In some embodiments, the virtual call generation module 220 may be configured to generate virtual proactive call content based on trigger conditions and to generate virtual call response content based on a passenger's response to the virtual proactive call content and / or environmental conditions.

[0044] In some embodiments, the virtual call generation module 220 can also be used to generate the text of the conversation based on pre-stored text fragments and conversation information through a lightweight model in the mobile terminal; and generate a virtual call voice based on the text of the conversation.

[0045] In some embodiments, the virtual call generation module 220 further includes a virtual incoming call reminder module ( Figure 2 (not shown). The virtual incoming call reminder module may be used to generate a virtual incoming call reminder based on a trigger condition.

[0046] In some embodiments, the virtual call system 200 may further include a service platform 230 .

[0047] The service platform 230 refers to a platform that enables data transmission and communication between the backend system and the frontend application, passengers and drivers. For example, by using technologies such as application programming interface (API), real-time communication, SMS and push notifications, databases and message queues, the service platform 230 can realize functions such as ride-hailing process, real-time data exchange and information push. Only when the service platform conditions are met can smooth communication be guaranteed.

[0048] In some embodiments, when the service platform conditions meet the requirements, the service platform 230 can generate virtual call response content through the large language model. For more information about the service platform conditions, please refer to Figure 4 .

[0049] In some embodiments, the virtual call system 200 may further include a judgment module 240. The judgment module 240 is used to start the recording or video recording software of the mobile terminal in response to the passenger's response to the virtual active call content meeting the first alarm condition; based on the recording or video content recorded by the recording or video recording software, determine whether the passenger's response to the virtual active call content meets the second alarm condition; in response to the passenger's response to the virtual active call content meeting the second alarm condition, send an alarm signal. In some embodiments, the judgment module 240 is used to determine whether the service platform conditions meet the requirements.

[0050] In some embodiments, the virtual call system 200 may further include a call quality detection module 250. The call quality detection module 250 is used to suspend the virtual call and / or simulate call quality changes based on whether the communication signal state and / or positioning information of the mobile terminal meets preset conditions.

[0051] Figure 3 It is an exemplary flow chart of a virtual call method according to some embodiments of the present specification.

[0052] like Figure 3 As shown, the process 300 includes the following steps. In some embodiments, the process 300 may be executed by the processor 110 .

[0053] Step 310: In response to a trigger condition, start a virtual call.

[0054] The trigger condition is a condition for determining whether the processor 110 can start a virtual call. In some embodiments, the trigger condition includes a passenger's trigger action and / or an environmental state.

[0055] The trigger action is the action or event that starts the virtual call. For example, the user makes a preset trigger action in the specified area. Exemplary, click / long press / drag the "virtual call" function area on the user terminal interaction interface, etc. In some embodiments, when the user is in the process of taking a taxi, the user can directly enter the "virtual call" function area based on the user's trigger action through the lock screen interface, making the virtual call more natural. The trigger action can be pre-set according to actual needs, and this manual does not limit this.

[0056] The environmental state refers to the environmental conditions of the vehicle during the ride, which can be reflected by the environmental parameters of the user during the ride. For example, the environmental parameters may include but are not limited to motion characteristics, route driving characteristics, and in-vehicle sound characteristics.

[0057] Motion characteristics refer to the characteristics exhibited by the vehicle that the user is riding in during driving, such as acceleration, angular velocity, etc.

[0058] Route driving characteristics refer to the characteristics of the route that the user is riding in. For example, the locations involved in the driving route and the time it takes to pass through, the deviation from the navigation route, and the remoteness of the locations involved in the route.

[0059] The in-vehicle sound features refer to the features related to the sound in the vehicle that the user is riding in. For example, the conversation frequency, volume, etc.

[0060] The triggering conditions can be pre-set based on actual needs. In some embodiments, in response to the user's triggering action (e.g., the user clicks on the "virtual call" function area on the terminal interaction interface), the processor 110 can control the virtual call of the terminal to be turned on. In some embodiments, in response to the environmental state satisfying the environmental conditions, the processor 110 can control the virtual call of the terminal to be turned on. Among them, the environmental conditions can include at least one of the following: (1) motion conditions: greater than the vehicle acceleration threshold, angular velocity threshold, etc.; (2) route conditions: greater than the deviation threshold, remoteness threshold, etc.; (3) sound conditions: greater than the frequency threshold, volume threshold, etc.

[0061] The virtual call refers to a call with a user simulated by the processor 110. For example, the object of the call with the user may be a telephone robot.

[0062] In some embodiments, the processor 110 may generate a virtual incoming call reminder based on a trigger condition.

[0063] Incoming call alert refers to the alert when the user receives an incoming call. For example, the screen displays the incoming call number, name and / or the incoming call alert tone of the contact.

[0064] In some embodiments, the processor 110 may generate different virtual incoming call reminders. For example, the processor 110 may generate different prompt tones and / or simulate different incoming call display interfaces.

[0065] In some embodiments, the virtual incoming call reminder includes a virtual incoming phone call and a virtual incoming call reminder sent by social software in the mobile terminal.

[0066] An incoming call refers to a call made through a SIM card. In some embodiments, the processor 110 may control a virtual call generation module to call the user using a real phone number.

[0067] The virtual incoming call reminder sent by the social software in the mobile terminal refers to the incoming call reminder that simulates a call through a third-party application. The third-party application may include but is not limited to QQ, WeChat, etc.

[0068] In some embodiments, the processor 110 may query a preset call table to determine a prompt tone and a caller ID interface based on a user input or random selection of a virtual phone call reminder or a virtual social software call reminder in a mobile terminal, based on the selected call reminder method. The preset call table stores prompt tones and caller ID interfaces corresponding to different call reminder methods.

[0069] In some embodiments, the virtual incoming call reminder may also include a preset incoming call reminder. The preset incoming call reminder includes at least a preset incoming call display interface and a preset call voice. The preset incoming call reminder may be pre-set and stored in the storage device 140. In some embodiments, the preset call voice may be generated based on the text of the conversation. For more information about generating a call voice based on the text of the conversation, see Figure 5 In some embodiments, in response to satisfying the trigger condition, the processor 110 may call a preset incoming call reminder, the terminal 130 may display a preset incoming call display interface and play a preset call voice, and a virtual call between the passenger and the terminal 130 may be performed based on playing the preset call voice.

[0070] In some embodiments of the present specification, generating different prompt tones or simulating different caller ID interfaces helps to ensure the rationality and naturalness of the virtual call.

[0071] Step 320: Generate virtual active call content.

[0072] Active call content refers to the content of the conversation that the robot actively initiates with the user at the beginning of the virtual call. For example, "How long will it take you to arrive?", "Where are you now?", etc. Active call content can also include any other content that can show concern for the user's car order.

[0073] In some embodiments, the processor 110 may generate active call content based on the user's car order. For example, the processor 110 generates active call content based on the user's boarding time and expected disembarkation time.

[0074] Step 330 , generating virtual call response content based on the passenger's response to the virtual active call content and / or the environmental state.

[0075] Call response content refers to the content of the robot's response to the user's answer during a virtual call.

[0076] In some embodiments, the processor 110 may generate virtual call response content based on the passenger's response to the virtual active call content / environmental status in combination with the real-time scenario.

[0077] In some embodiments, the processor 110 may generate call response content based on the vehicle route or the user's location information. For example, when the vehicle route or the user's location information includes a well-known scenic spot or building, the call response content may involve a discussion of the scenic spot. Exemplary call response content may include "Have you passed by XXX?", "Do you know the story about that place?", etc.

[0078] In some embodiments, the processor 110 may obtain hot spot information within a preset time period of the current date based on the current date, and generate call response content based on the hot spot information, for example, "Do you know the recent popular movie?" etc.

[0079] In some embodiments, the processor 110 can also generate virtual call response content through a large language model. For more information, see Figure 4 and related instructions.

[0080] In some embodiments, the active call content and the call response content are text data, and the processor 110 needs to generate corresponding virtual call voices based on the active call content and the call response content.

[0081] In some embodiments, the processor 110 can generate the text of the conversation based on the pre-stored text fragments and conversation information through a lightweight model, and generate a virtual call voice based on the text of the conversation. For more information, see Figure 5 and related instructions.

[0082] In some embodiments of the present specification, virtual active call content and call response content are generated based on environmental conditions, which helps to ensure the rationality and naturalness of the virtual call content, while indirectly warning people who may have bad intentions and avoiding causing disgust to people without bad intentions.

[0083] It should be noted that the above description of process 300 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0084] Figure 4 1 is an exemplary schematic diagram of a method for generating call response content according to some embodiments of this specification. In some embodiments, the method for generating call response content may be executed by the processor 110.

[0085] In some embodiments, the virtual call further includes: when the service platform conditions meet the requirements, obtaining virtual call response content generated by the large language model from the service platform.

[0086] like Figure 4 As shown, in some embodiments, the judgment module 240 can judge whether the service platform conditions meet the requirements 410, and in response, the processor 110 obtains the virtual call response content 430 generated by the large language model 420 from the service platform 230.

[0087] The service platform condition refers to the data processing capability of the service platform 230. In some embodiments, the service platform condition may include the service platform transmission capability and the service platform computing capability.

[0088] In some embodiments, the service platform transmission capability includes platform bandwidth. In some embodiments, the judgment module can judge whether the service platform condition meets the minimum platform bandwidth, and in response, the processor 110 obtains the virtual call response content generated by the large language model from the service platform. Among them, the minimum platform bandwidth refers to the minimum bandwidth that the service platform can support data transmission and real-time communication.

[0089] In some embodiments, the computing power of the service platform can be measured by requests per second (RPS) or throughput. A higher RPS or throughput means that the service platform can handle more requests at the same time and provide higher concurrency performance. In some embodiments, the judgment module can determine whether the service platform conditions meet the minimum RPS or throughput that can achieve a communication connection. In response, the processor 110 obtains the virtual call response content generated by the large language model from the service platform. In some embodiments, the computing power of the service platform can also be measured by parameters such as the time from the platform receiving the request to returning the response, the number of server nodes, and load balancing.

[0090] Large Language Model (LLM) refers to a machine learning model trained based on deep learning technology, large-scale data and computing resources. It is mainly used for natural language processing, but can also be used to process other forms of data. Typical large language models include GPT, BERT, etc.

[0091] In some embodiments, the large language model may be a conversation pattern-based large language model.

[0092] A large language model based on a conversational mode refers to a large language model with the following characteristics: the large language model is determined through pre-training and no fine-tuning is performed during the application process; the large language model is used in a conversational manner, and the input of the large language model may include the content to be processed and the context of the prompt; the conversation may include multiple rounds, and the content of the previous round of conversations can be used as a reference for subsequent output. Exemplary large language models include, but are not limited to, GPT and Bard.

[0093] In the large language model, the context of the prompt refers to a piece of text or question input to the large language model, which is used to guide the large language model to generate the corresponding output. For example, the prompt can be cleverly designed to guide the large language model to generate specific content, style or format when generating text. For example, the context of the prompt can be "Please generate a chat content related to the latest hot topic", "Speak more friendly", "Please recommend some popular movies", etc.

[0094] In some embodiments, the content to be processed by the large language model may include the passenger's response to the virtual active call content. In some embodiments, the content to be processed and the context of the prompt may be input into the large language model to generate the virtual call response content.

[0095] In some embodiments, the context of the prompt can be generated based on the processing of the content to be processed by the large language model; and the virtual call response content of the large language model can be generated based on the context of the prompt. For example, the content to be processed by the large language model can be the passenger's response "What are you going to eat for dinner today?", and the large language model generates the context of the prompt "Please generate chat content based on dinner." based on the passenger's response; the large language model generates the virtual call response content "I want to try spaghetti bolognese at a newly opened restaurant" based on the context of the prompt.

[0096] In some embodiments, the large language model generating virtual call response content also includes a text preprocessing step for the input content to be processed, for example, converting the conversation or user input into a form that the large language model can understand through preprocessing steps such as word segmentation, tokenization, and encoding.

[0097] In some embodiments, the generation of virtual call response content by the large language model may also include processing and optimizing the generated virtual call response content, including removing redundant information, correcting grammatical errors, adjusting the tone and style of the response, etc., to ensure its fluency, coherence and semantic correctness.

[0098] In some embodiments, the virtual call further includes sending current location information to the service platform 230 for context prompted by the large language model.

[0099] In some embodiments, the content to be processed by the large language model may include current location information. In some embodiments, the current location information and the context of the prompt may be input into the large language model to generate virtual call response content.

[0100] In some embodiments, the context of the prompt can be generated based on the processing of the current location information by the large language model; the virtual call response content of the large language model is generated based on the context of the prompt. For example, the content to be processed by the large language model can be the current location information "near the ancient castle", and the large language model generates the context of the prompt "Please generate chat content based on the ancient castle" based on the current location information; the large language model generates the virtual call response content "You should pass by a building that looks like a castle. Do you know what that is? That building is called an ancient castle. It was built several centuries ago. It has a long history and a unique architectural style."

[0101] The embodiment of this specification generates a virtual call response content and combines it with the current location information of the real-time scene, and uses the current location information in the context of the large language model prompt, making the virtual call more realistic and natural.

[0102] In some embodiments, the call response content is text, and the mobile terminal generates voice based on the text, and simulates the call quality change when generating the voice. For more information, see Figure 5 Related instructions.

[0103] Some embodiments of the present specification generate more complex and natural sentences through a large language model, and simulate call quality when generating response content, thereby more intelligently improving the quality and fluency of the conversation.

[0104] Figure 5 is an exemplary schematic diagram of a lightweight model according to some embodiments of the present specification.

[0105] In some embodiments, the processor 110 may generate a text 560 of a conversation based on a pre-stored text segment 510 and conversation information 520 through a lightweight model 550 in the mobile terminal; and generate a virtual call voice 570 based on the text 560 of the conversation.

[0106] The pre-stored text segment is text data pre-stored in the virtual call system. For example, the pre-stored text segment may include "Have you watched XXX program recently?", "What do you think of XXX?", etc.

[0107] In some embodiments, processor 110 may update the pre-stored text segments based on news data of the current date.

[0108] The conversation information refers to various information related to the virtual call. For example, the conversation information may include the content of the active call, the content of the call response, and the environment status, etc. For details, please refer to the relevant description above.

[0109] The text of the conversation refers to the textual data of the virtual conversation content.

[0110] A lightweight model refers to a model that has smaller computing resource requirements than traditional deep learning models.

[0111] The input of the lightweight model may include pre-stored text segments 510 and dialogue information 520 .

[0112] In some embodiments, the input of the lightweight model may also include the time 530 of the virtual call and the location information 540 of the passengers.

[0113] The location information may include at least the user's current location and the locations along the driving route, which may be determined through a network-accessible positioning system or based on the driving route.

[0114] The location information may also include the user's future location and order destination. In some embodiments, the processor 110 may input the user's current location, departure time, order starting point, and order destination into a location determination model to predict the user's future location. The user's future location refers to the location that the user will pass through. The departure time refers to the time it takes the user to get from the order starting point to the current location.

[0115] The location determination model refers to a machine learning model used to predict the user's future location. In some embodiments, the location determination model can be a neural network model (NN), a deep neural network (DNN), a convolutional neural network (CNN), etc. or any combination thereof.

[0116] In some embodiments, the processor 110 may train a location determination model based on a plurality of first training samples with a first label. The first training sample may include the location of the sample user at the first historical time point, the sample departure time, the sample order starting point, and the sample order destination, and the first label is the location of the sample user at the second historical time point. In some embodiments, the first training sample may be obtained based on historical data, and the first label may be determined based on manual annotation. For example, a landmark or scenic spot between the location of the sample user at the first historical time point and the sample order destination may be determined as the first label. The first historical time point is before the second historical time point.

[0117] In an exemplary training process, the location of the sample user at the first historical time point, the sample departure time, the sample order starting point, and the sample order destination are input into the initial location determination model, and the location of the sample user at the second historical time point output by the initial location determination model is obtained. During the training process, the initial location determination model can construct a loss function based on the first label and the output of the initial location determination model, and update the parameters of the initial location determination model until the preset conditions are met and the training is completed. The preset conditions may include one or more of the loss function being less than a threshold, convergence, or the training cycle reaching a threshold.

[0118] In some embodiments, the processor 110 may also generate text of a conversation related to the user's future location based on the user's future location generated by the location determination model through the lightweight model 550 .

[0119] The time of the virtual call may include the current time and date, which may be obtained by accessing a clock or other feasible means.

[0120] In some embodiments, the lightweight model 550 can be trained by the training sample set 590. The training sample set includes a second training sample and a second label. The second training sample may include a sample pre-stored text segment, sample conversation information, the time of a sample virtual call, and the location information of a sample passenger. The second label is the sample call response content generated by the large language model. For example, multiple second training samples of the training sample set 590 can be input into the initial lightweight model 580, and a loss function can be constructed by the second label and the result of the initial lightweight model 580, and the parameters of the initial lightweight model 580 can be iteratively updated based on the loss function. When the loss function of the initial lightweight model 580 meets the iterative preset conditions, the model training is completed, and the trained lightweight model 550 is obtained. Among them, the iterative preset conditions can be that the loss function converges, the number of iterations reaches a threshold, etc.

[0121] In some embodiments of the present specification, by generating and automatically updating conversation text packages through a lightweight model, the storage space, memory usage, and calculation time of the model can be reduced while maintaining the highest possible accuracy and effect, thereby helping the model to have better adaptability and usability in resource-constrained environments (such as mobile devices) and avoid the impact of poor signals.

[0122] The call voice refers to the voice data of the virtual call, for example, the voice corresponding to the virtual active call content, the voice corresponding to the call response content, etc.

[0123] In some embodiments, the processor 110 may generate a virtual call voice 570 based on the conversation text 560 by a preset method. The preset method may include any text-to-speech method, which is not limited in this specification.

[0124] In some embodiments, based on the communication signal status and / or positioning information of the mobile terminal satisfying preset conditions, the processor 110 may suspend the virtual call and / or simulate call quality changes to improve the authenticity of the virtual call.

[0125] The communication signal state refers to the communication information exchange and transmission situation. When the communication signal state is poor, the communication information may not be exchanged smoothly or cannot be transmitted. In some embodiments, the communication signal state can be determined by a signal monitoring device (eg, a modem) of the mobile terminal.

[0126] The preset condition refers to a condition for determining whether to suspend a virtual call and / or change call quality. In some embodiments, the preset condition may include a communication signal condition. For example, the communication signal condition may include a communication signal threshold.

[0127] In some embodiments, the preset condition may also include a preset location condition. For example, the preset location condition may include that the passenger is in a tunnel, suburb, or other place with poor communication signal. Figure 8 As shown, in some embodiments, in response to the communication signal state being less than the communication signal threshold, or in response to the location information of the mobile terminal being in a tunnel, suburbs, or other place where the communication signal is poor or people generally believe that the signal is poor, the processor 110 can control the mobile terminal to suspend the virtual call and / or reduce the call quality. In some embodiments, if the virtual call is suspended, the processor 110 can control the mobile terminal to issue a voice prompt "The user you dialed is temporarily unavailable, please call again later". In some embodiments, the processor 110 can change the call quality by changing the voice clarity, audio delay, jitter, etc.

[0128] In some embodiments of the present specification, the authenticity of the virtual call can be improved by pausing the virtual call and / or simulating the change in call quality through the relationship between the communication signal status or positioning information and the preset conditions. Although the communication information exchange and transmission situation can also be judged based on the communication signal status alone, combining the location information is more helpful to improve the rationality and authenticity, and avoid calls in places where the signal is poor or people generally believe that the signal is poor, which may arouse suspicion from people in the car.

[0129] Figure 6 It is an exemplary schematic diagram of determining whether to issue an alarm signal according to some embodiments of this specification.

[0130] In some embodiments, in response to the virtual call being started, the processor 110 starts the recording software of the mobile terminal to obtain the content of the virtual call. For example, when the passenger clicks the "virtual call" function area on the mobile terminal interactive interface, the recording software of the mobile terminal is started synchronously. In some embodiments, the judgment module can determine whether to alarm based on the content of the recording. For example, if a recording of a voice calling for help or a nervous tone is recognized, an alarm can be issued.

[0131] In some embodiments, in response to the passenger's response to the virtual active call content satisfying the first alarm condition 610, the processor 110 can activate the recording or video recording software 620 of the mobile terminal.

[0132] The first alarm condition refers to a condition for determining whether to activate the recording or video recording software. In some embodiments, the first alarm condition may include the passenger actively activating the recording or video recording function. In some embodiments, the first alarm condition may also include the passenger's response to the active call content including dangerous keywords. Dangerous keywords may be pre-set.

[0133] In some embodiments, the passenger responds to the virtual active call content according to the guidance or prompts of the mobile terminal interface. Figure 7 As shown, the virtual active call content is "Would you like to eat pizza or pasta?" The processor 110 can guide and prompt the passenger through the interactive interface, for example, "Please say pizza if it is dangerous, and please say pasta if it is safe", where pizza is a pre-set danger keyword, which can indicate that the passenger thinks that the environment he is in is dangerous. If the passenger responds with "Let's eat pizza", the judgment module can determine that the passenger's response to the virtual active call content meets the first alarm condition.

[0134] In some embodiments of the present specification, passengers are guided to respond through the mobile terminal interface to improve the effectiveness of the call, and dangerous keywords are set to help passengers further monitor the environmental status and relieve tension, while avoiding adverse effects on people without bad intentions.

[0135] In some embodiments, the passenger can actively start the recording or video recording software during the virtual call, or the judgment module recognizes that the passenger's response includes dangerous keywords and issues a control instruction to control the mobile terminal to start the recording or video recording software.

[0136] In some embodiments, the judgment module can judge whether the passenger's response to the virtual active call content meets the second alarm condition 640 based on the audio or video content 630 recorded by the audio or video software, and issue an alarm signal 650 in response to the passenger's response to the virtual active call content meeting the second alarm condition.

[0137] The second alarm condition refers to a condition for determining whether to alarm. In some embodiments, the second alarm condition may also include the passenger's response to the active call content including an alarm keyword. The alarm keyword may be pre-set.

[0138] The processor 110 can process the audio or video content in a variety of ways. In some embodiments, the processor 110 can process the audio or video content by voice recognition, obtain the corresponding text information, and determine whether to send an alarm signal based on whether the alarm keyword is included in the text information. The alarm keywords and danger keywords used in this article can both be used to determine the user's degree of danger. Among them, the user's degree of danger reflected by the alarm keyword is higher than the user's degree of danger reflected by the danger keyword. In some embodiments, the passenger's response to the virtual active call content is based on the guidance or prompt of the mobile terminal interface. For example, the virtual active call content is "Do you want to drink juice or wine later?" The processor 110 can guide and prompt the passenger through the interactive interface, for example, "Please say wine if you want to call the police, and please say juice if you don't want to call the police", where wine is a pre-set danger keyword, which can indicate that the passenger believes that the environment he is in is dangerous and needs to call the police. If the passenger responds to "drink wine", the judgment module can determine that the passenger's response to the virtual active call content meets the first alarm condition.

[0139] In some embodiments, the judgment module recognizes that the passenger's response includes dangerous keywords, and issues a control instruction to control the mobile terminal to start the recording or video recording software. In some embodiments, the judgment module can call a pre-defined alarm keyword list, which contains words related to dangerous situations, such as "help". The judgment module will match the passenger's response with these keywords to detect whether there are potential risk factors. In some embodiments, the alarm module can comprehensively include factors such as the frequency, context, and tone of the alarm keywords to determine whether the passenger's response truly represents a potential danger or emergency.

[0140] In some embodiments, in response to a user request for an alarm, the processor 110 sends an alarm signal, contacts customer service or security experts to confirm the audio or video content, and then sends an alarm after confirmation.

[0141] In some embodiments of the present specification, by setting the first alarm condition and the second alarm condition, when the passenger thinks there is danger but does not need to call the police, it can help the passenger monitor the riding environment, and if there is a dangerous situation, the police can be called in time to relieve the passenger's tension; when the user determines that an alarm is needed, the security expert can be contacted for confirmation to ensure the safety of the passengers and reduce the possibility of false alarms.

[0142] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0143] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "one embodiment", "an embodiment", and / or "some embodiments" refer to a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "one embodiment" or "an embodiment" or "an alternative embodiment" mentioned twice or more in different positions in this specification does not necessarily refer to the same embodiment. In addition, certain features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0144] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some invention embodiments that are currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0145] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0146] In some embodiments, numbers describing the number of components and attributes are used. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the setting of such numerical values ​​is as accurate as possible within the feasible range.

[0147] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, documents, etc., cited in this specification are hereby incorporated by reference in their entirety. Except for application history documents that are inconsistent with or conflicting with the contents of this specification, documents that limit the broadest scope of the claims of this specification (currently or later attached to this specification) are also excluded. It should be noted that if the descriptions, definitions, and / or use of terms in the materials attached to this specification are inconsistent or conflicting with the contents described in this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0148] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A virtual call method, the method being executed by a mobile terminal, comprising: In response to a trigger condition, a virtual call is initiated; The triggering condition includes a passenger's triggering action and / or an environmental state; The virtual call includes: Generate virtual active call content; Based on the passenger's response to the virtual active call content and / or the environmental state, a virtual call response content is generated.

2. The method according to claim 1, characterized in that The virtual call includes: Based on the trigger condition, a virtual incoming call reminder is generated.

3. The method according to claim 1, characterized in that The virtual call further includes: when the service platform conditions meet the requirements, obtaining the virtual call response content generated by the large language model from the service platform.

4. The method according to claim 1, characterized in that The virtual call also includes: Based on the pre-stored text fragments and the conversation information, the text of the conversation is generated by the lightweight model in the mobile terminal; A virtual conversation voice is generated based on the text of the conversation.

5. The method according to claim 1, characterized in that Also includes: In response to the passenger's response to the virtual active call content meeting a first alarm condition, starting a recording or video recording software of the mobile terminal; Based on the audio or video content recorded by the audio or video recording software, determining whether the passenger's response to the virtual active call content meets a second alarm condition; In response to the passenger's response to the virtual active call content satisfying a second alarm condition, an alarm signal is issued.

6. The method according to claim 1, characterized in that Also includes: Based on the communication signal status and / or positioning information of the mobile terminal meeting a preset condition, the virtual call and / or simulated call quality change is suspended to improve the authenticity of the virtual call.

7. A virtual call system, characterized in that: include: A detection module, used to detect a triggering condition for a virtual call, wherein the triggering condition includes a triggering action of a passenger and / or an environmental state; The virtual call generation module is used to generate virtual active call content based on the trigger condition and generate virtual call response content based on the passenger's response to the virtual active call content and / or the environmental state.

8. The system according to claim 7, characterized in that The virtual call generation module further includes a virtual incoming call reminder module, and the virtual call generation module is further used for: Based on the trigger condition, a virtual incoming call reminder is generated.

9. The system according to claim 7, characterized in that Also includes: The service platform is configured to generate the virtual call response content through a large language model when platform conditions meet the requirements.

10. The system according to claim 7, characterized in that The virtual call generation module is also used for: Based on the pre-stored text fragments and the conversation information, the text of the conversation is generated by the lightweight model in the mobile terminal; A virtual conversation voice is generated based on the text of the conversation.

11. The system according to claim 7, characterized in that It also includes a judgment module, which is used to: In response to the passenger's response to the virtual active call content meeting a first alarm condition, starting a recording or video recording software of the mobile terminal; Based on the audio or video content recorded by the audio or video recording software, determining whether the passenger's response to the virtual active call content meets a second alarm condition; In response to the passenger's response to the virtual active call content satisfying a second alarm condition, an alarm signal is issued.

12. The system according to claim 7, characterized in that It also includes a call quality detection module, which includes: Based on the communication signal status and / or positioning information of the mobile terminal meeting a preset condition, the virtual call and / or simulated call quality change is suspended.

13. A virtual communication device, characterized in that: The apparatus comprises at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least part of the computer instructions to implement the method according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions, and when the computer instructions are read by a computer, the method according to any one of claims 1 to 6 is executed.

Images