CN115118820A - Call processing method and device, computer equipment and storage medium - Google Patents

Abstract

The embodiments of the present application disclose a call processing method, device, computer equipment, and storage medium. This solution obtains historical call records of a target user; based on a preset dual-modal neural network model, the voice information and The text information is extracted separately to obtain the speech features and text features of the target users when they face various types of discourse templates; the preference degree of each type of discourse template; based on the preference degree of the target user when faced with various types of discourse templates, the first target discourse template with the highest preference is determined from the various types of discourse templates, and the first target discourse template with the highest preference is determined. The target speech template performs a prompt operation, so that the customer service agent can communicate with the target user according to the first target speech template, thereby improving the service quality of the telephone agent service.

Description

A call processing method, device, computer equipment and storage medium

technical field

The present application relates to the field of call technology, and in particular, to a call processing method, device, computer equipment, and storage medium.

Background technique

With the development of financial technology and social economy, banks have specially set up telephone agent services to better serve customers. The existing telephone agent service mainly relies on the experience of the customer service agent to provide services, and the service quality of the telephone agent service is poor due to the large mobility of the customer service agent and the uneven professional level.

SUMMARY OF THE INVENTION

The embodiments of the present application provide a call processing method, apparatus, computer equipment and storage medium, which can improve the service quality of the telephone agent service.

The embodiment of the present application provides a call processing method, including:

Obtain the historical call records of the target user, the above-mentioned historical call records include the call records when the above-mentioned target user is called based on various types of speech templates;

Based on a preset dual-modal neural network model, feature extraction is performed on the voice information and text information corresponding to the above-mentioned historical call records, respectively, to obtain the above-mentioned target user's voice features and text features when faced with various types of discourse templates;

Determine the degree of preference of the above-mentioned target user when facing various types of discourse templates according to the speech features and text characteristics of the above-mentioned target users when they face various types of discourse templates;

Based on the degree of preference of the target user when faced with various types of speech templates, the first target speech template with the highest preference is determined from the above various types of speech templates, and the above-mentioned first target speech template is prompted. , so that the customer service agent can communicate with the target user according to the first target speech template.

Correspondingly, an embodiment of the present application also provides a call processing device, including:

A record acquisition module, used for acquiring the historical call records of the target user, the above-mentioned historical call records including the call records when the above-mentioned target user is called based on various types of speech templates;

The feature extraction module is configured to perform feature extraction on the voice information and text information corresponding to the above-mentioned historical call records based on a preset dual-modal neural network model, respectively, to obtain the above-mentioned target user's voice characteristics when faced with various types of discourse templates and text features;

A degree determination module, configured to determine the degree of preference of the above-mentioned target user when facing various types of discourse templates according to the speech features and text characteristics of the above-mentioned target users when they face various types of discourse templates;

The prompting operation module is used to determine the first target speech template with the highest degree of preference from the above-mentioned various types of speech templates based on the preference degree of the target user when facing various types of speech templates, and assign the above-mentioned first target speech template. The speech template performs a prompt operation, so that the customer service agent can communicate with the target user according to the first target speech template.

In some embodiments, the above-mentioned call processing apparatus further includes:

a first information acquisition module, configured to acquire attribute information of the target user when the target user is a new user;

a user determination module, configured to determine historical users whose attributes are similar to the above-mentioned target user according to the attribute information of the above-mentioned target user;

The template obtaining module is used to obtain the second target speech template with the highest preference corresponding to the above-mentioned historical users, and perform a prompt operation on the above-mentioned second target speech template, so that the above-mentioned customer service agent can match the second target speech template with the above-mentioned second target speech template. The above target user makes a call.

In some embodiments, the above-mentioned call processing apparatus further includes:

a second information acquisition module, configured to acquire call information within a preset time period of the target user when the target user is a new user;

a scenario determination module, configured to determine the business scenario required by the target user according to the call information;

The template determination module is configured to determine the third target speech template corresponding to the above-mentioned business scenario according to the above-mentioned business scenario, and perform a prompt operation on the above-mentioned third target speech template, so that the above-mentioned customer service agent can match the above-mentioned third target speech template with the above-mentioned third target speech template. The above target user makes a call.

In some embodiments, the above-mentioned call processing apparatus further includes:

a prediction module, configured to acquire call records during the call in real time, input the call records into a preset neural network model to predict the preference degree, and obtain the preference degree of the target user for the first target speech template;

The selection module is configured to select a fourth target speech template from the preset template set according to the call record when the preference degree is less than a preset threshold, and perform a prompt operation on the fourth target speech template, so that the customer service The agent communicates with the target user according to the fourth target speech template.

In some embodiments, the above-mentioned selection module includes:

a scenario determination unit, configured to determine the business scenario currently required by the target user according to the call record;

The first selection unit is configured to select the fourth target speech template corresponding to the above-mentioned business scenario from the above-mentioned template set.

In some embodiments, the above-mentioned first target speech template includes at least one service phase, the above-mentioned service phase corresponds to at least one speech sub-template, and the above-mentioned call processing apparatus further includes:

a stage determination module, used for obtaining the current reply information of the target user, and determining the current business stage according to the above-mentioned current reply information;

A highlighting and prompting module is used to select a target sub-template from at least one speech sub-template corresponding to the above-mentioned current business stage, and to highlight and prompt the above-mentioned target sub-template.

In some embodiments, the above-mentioned highlighting and prompting module includes:

a state acquisition unit, used for acquiring the emotional state of the above-mentioned target user;

The second selecting unit is configured to select a target sub-template from at least one speech sub-template corresponding to the current business stage according to the emotional state.

Correspondingly, an embodiment of the present application also provides a computer device, including a memory, a processor, and a computer program stored on the memory and running on the processor, wherein the processor executes any of the provided embodiments of the present application. Call handling method.

Correspondingly, an embodiment of the present application further provides a storage medium, where the storage medium stores a plurality of instructions, and the instructions are suitable for the processor to load, so as to execute the above call processing method.

This embodiment of the present application acquires the historical call records of the target user, and the above-mentioned historical call records include the call records when the above-mentioned target users are called based on various types of discourse templates; The voice information and text information corresponding to the call records are extracted respectively to obtain the voice features and text features of the above-mentioned target users when they are faced with various types of discourse templates; and text features to determine the degree of preference of the target user when faced with various types of discourse templates; based on the degree of preference of the target user when faced with various types of discourse templates, the highest preference degree is determined from the above various types of discourse templates the first target speech template, and perform a prompt operation on the above-mentioned first target speech template, so that the customer service agent can communicate with the above-mentioned target user according to the above-mentioned first target speech template, so as to determine the historical call records of the target user. The target user's most preferred speech template, and then based on the speech template to prompt the telephone agent service to communicate with the target user, the professional level of the telephone agent service and the service quality of the telephone agent service are improved, and the user's call experience is improved.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

FIG. 1 is a schematic flowchart of a call processing method provided by an embodiment of the present application.

FIG. 2 is a structural block diagram of a call processing apparatus provided by an embodiment of the present application.

FIG. 3 is a schematic structural diagram of a computer device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

Embodiments of the present application provide a call processing method, apparatus, storage medium, and computer equipment. Specifically, the call processing method in the embodiment of the present application may be executed by a computer device, where the computer device may be a server, or may be a terminal or other device. The server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or may provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, and cloud communications. , middleware services, domain name services, security services, CDN, and cloud servers for basic cloud computing services such as big data and artificial intelligence platforms. The terminal may be a smart phone, a desktop computer, a notebook computer, a tablet computer, etc., but is not limited thereto. The terminal and the server may be directly or indirectly connected through wired or wireless communication, which is not limited in this application.

For example, the computer device may be a terminal, and the terminal may acquire historical call records of the target user, and the historical call records include call records when the target user is called based on various types of speech templates; state neural network model, respectively perform feature extraction on the voice information and text information corresponding to the above-mentioned historical call records, and obtain the voice and text characteristics of the above-mentioned target users when they face various types of discourse templates; The speech features and text features of the types of discourse templates determine the degree of preference of the above-mentioned target users when facing various types of discourse templates; From the speech templates, determine the first target speech template with the highest degree of preference, and perform a prompt operation on the first target speech template, so that the customer service agent can communicate with the target user according to the first target speech template.

Based on the above problems, the embodiments of the present application provide a first call processing method, device, computer equipment, and storage medium, which can improve the service quality of the telephone agent service.

Each of them will be described in detail below. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

The embodiment of the present application provides a call processing method, and the method can be executed by a terminal or a server. The embodiment of the present application takes the call processing method executed by a terminal as an example for description.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart of a call processing method provided by an embodiment of the present application. The specific flow of the call processing method may be as follows:

101. Obtain the historical call records of the target user.

In this embodiment, the above-mentioned historical call record is a call record generated when a customer service agent makes a call to a target user based on various types of discourse templates in historical time. Call records when the user makes a call.

Among them, the above-mentioned customer service agents are the telephone customer service personnel who talk with the user; the above-mentioned speech template includes but is not limited to the relevant content of greeting, the relevant content of replying to the user's question, and the relevant content of asking the user to wait, etc.; Types include, but are not limited to, concise type, gentle type, cute type, etc.

102. Based on a preset dual-modal neural network model, feature extraction is performed on the voice information and text information corresponding to the historical call records, respectively, to obtain the voice features and text features of the target user when faced with various types of discourse templates.

In this embodiment, the bimodal neural network model is obtained by training the voice data and text data in advance. When there is a business requirement, that is, when the current customer service agent needs to talk with the target user, the terminal can use the bimodal neural network model to The acquired historical call records of the target user are subjected to feature extraction of two modalities, and the two modal decibels are two modalities of speech and text. Phonetic and textual features for each particular type of utterance template.

It is understandable that, in general, historical call records include customer service voice and user voice. Since the current feature extraction is mainly for user voice, user voice can be extracted from historical call records, so as to perform feature extraction on user voice. And when the feature extraction of the text is carried out, the user's voice can be converted into text information in advance, and then the voice information corresponding to the user's voice and the text information after the user's voice conversion are input into the preset bimodal neural network model for feature extraction. .

103. Determine, according to the speech features and text features of the target user when the target user faces various types of utterance templates, the degree of preference of the target user when faced with various types of utterance templates.

In this embodiment, since the target users face different types of discourse templates, their preference degrees are also different, so after obtaining the speech features and text characteristics of the target users when they face each specific type of discourse template, it can be based on The bimodal neural network model performs preference degree recognition on the extracted speech features and text features, that is, the preference degree of the target user when faced with each specific type of discourse template.

For example, if the historical call records of the current target user include three types of speech templates, namely concise type speech templates, gentle type speech templates, and cute type speech templates, based on the dual-modal neural network model, the target The user's preference degree when facing the concise type of discourse template is A, the target user's preference degree when facing the gentle type discourse template is B, and the target user's preference degree when facing the cute type of discourse template is C.

104. Based on the preference degree of the target user when facing various types of speech templates, determine the first target speech template with the highest degree of preference from the various types of speech templates, and perform a prompt operation on the first target speech template, So that the customer service agent can communicate with the target user according to the first target speech template.

In this embodiment, by selecting from various types of speech templates in the historical call records of the target user, a speech template with a higher degree of relative preference of the target user, that is, the first target speech template, can be used as the first target speech template. The speech template is displayed on the touch screen of the terminal, prompting the customer service agent to communicate with the target user according to the first target speech template displayed on the display screen, so that the target user can be informed by using the speech template with the highest preference of the target user. Make calls to improve the service quality of customer service agents.

For example, based on the above example, it is further limited that the above B is greater than C and greater than A, so that the first target speech template with a higher relative preference of the target user can be obtained as a gentle type speech template, and then the terminal prompts the gentle type speech template Operation, so that the customer service agent can talk to the target user according to the gentle type speech template.

In some embodiments, due to different business scenarios corresponding to different discourse templates, there may be a phenomenon that at least two discourse templates of the same type correspond to different business scenarios, and the contents of discourse templates for different business scenarios also include The difference is that, in order to better improve the service quality of the telephone agent service, after performing the prompt operation on the first target speech template, it may further include further real-time detection of the call records between the customer service agent and the target user, so as to judge in real time what the current target user has. Whether the required business scenario is consistent with the business scenario corresponding to the first speech template obtained before the communication, so that when the two business scenarios are inconsistent, it can be dealt with in time to improve the service quality of the telephone agent service. Among them, the above-mentioned response methods when the business scenarios are inconsistent include but are not limited to replacing the first target speech template, modifying the first target speech template, etc. The above business scenarios include but are not limited to after-sales scenarios and consulting scenarios. and sales scenarios.

Specifically, the terminal may acquire the call records of the customer service agent during the call process with the target user based on the first speech template in real time, and input the call records into a preset neural network model to predict the preference degree in real time, so as to obtain the current preference of the target user. The preference degree of the first target speech template. When the target user's current preference for the first target speech template is greater than or equal to the preset threshold, it means that the target user is still satisfied with the first target speech template. The first target speech template performs any processing; when the target user's current preference for the first target speech template is less than the preset threshold, it means that the target user is currently dissatisfied with the first target speech template, and both may To conduct unfriendly behaviors such as customer complaints, you can select the fourth target speech template from the preset template set according to the call record, and perform a prompt operation on the fourth target speech template, so that the customer service agent can use the fourth target speech template according to the fourth target speech. Use the technology template to talk to the target user.

Wherein, the type of the above-mentioned fourth target speech template may be consistent with the type of the first target speech template, but the business scenarios of the two are different.

The above-mentioned selecting the fourth target speech template from the preset template set according to the call record may specifically include: the terminal determines the business scenario currently required by the target user according to the call record; selecting the fourth target speech corresponding to the business scenario from the template set technique template. The template set can be determined according to the type of the first target vocabulary template, that is, the vocabulary templates of the same type and business scenarios can be stored in the same template set, and when there is a demand, according to the type of the first target vocabulary template Determine the template set corresponding to the type, and then select the fourth target language template corresponding to the business scenario from the determined template set, the fourth target language template is of the same type as the first target language template, and the fourth target language template is of the same type. The business scenarios of technology templates are more in line with the current needs of target users.

In some embodiments, since there are different stages in different business scenarios, for example, the greeting stage, the replying user question stage, and the user waiting stage, etc., the above-mentioned first target speech template applied to a specific business scenario also includes at least A business stage, this business stage corresponds to at least one speech sub-template, that is, a sub-template of the same type but different speech modes. In order to ensure service quality, after prompting the first target speech template, you can also include : The terminal obtains the current reply information of the target user, which is equivalent to the reply content of the user's reply to the customer service's words in the previous stage, so that it can determine which business stage the current communication is in the business scenario according to the current reply information, that is, to determine the current In the business stage, since the business stage corresponds to at least one conversation sub-template, the target sub-template can be selected from the at least one conversation sub-template corresponding to the current business stage, and the target sub-template can be highlighted on the touch screen. .

The above-mentioned selecting the target sub-template from at least one speech sub-template corresponding to the current business stage may specifically include: selecting the target sub-template from at least one speech sub-template corresponding to the current business stage by random selection, or using a specific rule to select the target sub-template. The target sub-template is selected from at least one speech sub-template corresponding to the current business stage. The specific rule may be determined according to the current state or attribute information of the target user, and the current state may be an emotional state; the attribute information may be the target user's gender, age, occupation and other information.

Specifically, selecting the target sub-template from at least one vocabulary sub-template corresponding to the current business stage according to the current state may include: the terminal may acquire the emotional state of the target user, and then select the target sub-template from at least one vocabulary corresponding to the current business stage according to the emotional state. Select the target sub-template from the sub-template. Wherein, the emotional state of the target user can be obtained by real-time acquisition of the call record of the customer service agent during the call with the target user based on the first speech template, so as to identify the emotional state of the target user according to the call record.

In some embodiments, since some target users may be new users, and the target users have no historical call records, when the target user is a new user, it is impossible to obtain the historical call records of the target user, and according to The historical call record determines the first target speech template with the highest preference of the target user, so when the target user is a new user, the target can be determined according to the attribute information of the target user or the call information within the preset time period of the target user The discourse template with the highest user preference. The above attribute information may be information such as gender, age, occupation, etc. of the target user.

It is understandable that, if the target user has historical records, but the first target vocabulary template cannot be determined according to the historical records, it means that although the target user has historical call records, different types of vocabulary templates in the historical call records correspond to different types of vocabulary templates. If the preference degree of the target users corresponding to the call records is too low, so that the first target speech template cannot be obtained according to the speech template with the too low preference degree, the target user can be regarded as a new user.

Specifically, it may include: when the target user is a new user, the terminal obtains attribute information of the target user, so as to determine historical users with similar attributes to the target user according to the attribute information of the target user, so as to obtain the second target corresponding to the historical user with the highest degree of preference The language template, and prompting the second target language template, so that the customer service agent can communicate with the target user according to the second target language template.

The above-mentioned historical user includes at least one, and for the acquisition method of the second target speech template with the highest degree of preference corresponding to the historical user, reference may be made to the above-mentioned steps 101 to 104 . When there are at least two historical users, according to the preference degrees of the respective types of discourse templates corresponding to the at least two historical users, a discourse template with the highest preference degree is selected from the discourse templates of various types corresponding to the at least two historical users respectively, That is, the second target speech template.

Specifically, it may include: when the target user is a new user, the terminal obtains call information within a preset time period of the target user, so as to determine the service scenario required by the target user according to the call information, thereby determining the third corresponding to the service scenario according to the service scenario. target language template, and perform a prompt operation on the third target language template, so that the customer service agent can communicate with the target user according to the third target language template.

Wherein, the above business scenario corresponds to at least one type of speech template, and a third target speech template may be selected from at least one type of speech template by random selection. The speech template of the target type with the largest number of people can also be determined as the third target speech template according to the number of people whose preference degree corresponding to each type of speech template is greater than the preset degree threshold.

In the embodiment of the present invention, "first", "second", "third", and "fourth" are only used to distinguish vocabulary templates, and have no other special meanings.

The embodiment of the present application discloses a call processing method. The method includes: acquiring historical call records of a target user, where the historical call records include call records when making calls to the target user based on various types of speech templates; The set dual-modal neural network model performs feature extraction on the voice information and text information corresponding to the above-mentioned historical call records, respectively, to obtain the above-mentioned target users when they face various types of discourse templates. The speech features and text features when faced with various types of discourse templates determine the degree of preference of the above target users when faced with various types of discourse templates; Among the above various types of speech templates, determine the first target speech template with the highest degree of preference, and perform a prompt operation on the above-mentioned first target speech template, so that the customer service agent can communicate with the above-mentioned target user according to the above-mentioned first target speech template. Therefore, the service quality of the telephone agent service can be improved.

In order to facilitate better implementation of the call processing method provided by the embodiment of the present application, the embodiment of the present application further provides a call processing apparatus based on the above-mentioned call processing method. The meanings of the nouns are the same as those in the above-mentioned call processing method, and the specific implementation details may refer to the descriptions in the method embodiments.

Please refer to FIG. 2. FIG. 2 is a structural block diagram of a call processing apparatus provided by an embodiment of the present application. The apparatus includes:

The record obtaining module 201 is configured to obtain the historical call records of the target user, and the historical call records include the call records when the target user is called based on various types of discourse templates.

The feature extraction module 202 is configured to perform feature extraction respectively on the voice information and text information corresponding to the historical call records based on the preset bimodal neural network model, and obtain the voice features and text features.

The degree determination module 203 is configured to determine the degree of preference of the target user when facing various types of discourse templates according to the speech features and text characteristics of the target user when facing various types of discourse templates.

The prompt operation module 204 is configured to, based on the preference degree of the target user when facing various types of speech templates, determine the first target speech template with the highest preference from the various types of speech templates, and assign the first target speech template to the first target speech template. The template performs a prompt operation, so that the customer service agent can communicate with the target user according to the first target speech template.

In some embodiments, the above-mentioned call processing apparatus further includes:

The first information acquisition module is configured to acquire attribute information of the target user when the target user is a new user.

The user determination module is used for determining historical users with attributes similar to the target user according to the attribute information of the target user.

The template acquisition module is used to obtain the second target speech template with the highest preference corresponding to the historical user, and perform a prompt operation on the second target speech template, so that the customer service agent can communicate with the target user according to the second target speech template .

In some embodiments, the above-mentioned call processing apparatus further includes:

The second information obtaining module is configured to obtain call information of the target user within a preset time period when the target user is a new user.

The scenario determination module is used to determine the service scenario required by the target user according to the call information.

The template determination module is used to determine the third target speech template corresponding to the business scenario according to the business scenario, and perform a prompt operation on the third target speech template, so that the customer service agent can communicate with the target user according to the third target speech template.

In some embodiments, the above-mentioned call processing apparatus further includes:

The prediction module is used to obtain the call records during the call in real time, input the call records into the preset neural network model to predict the preference degree, and obtain the preference degree of the target user for the first target speech template.

The selection module is configured to select the fourth target speech template from the preset template set according to the call record when the preference degree is less than the preset threshold, and perform a prompt operation on the fourth target speech template, so that the customer service agent can perform a prompt operation according to the fourth target speech template. The target speech template communicates with the target user.

In some embodiments, the above-mentioned selection module includes:

The scenario determination unit is used to determine the current business scenario required by the target user according to the call record.

The first selection unit is configured to select the fourth target speech template corresponding to the business scenario from the template set.

In some embodiments, the first target speech template includes at least one service phase, and the service phase corresponds to at least one speech sub-template, and the above-mentioned call processing apparatus further includes:

The stage determination module is used for acquiring the current reply information of the target user, and determining the current business stage according to the current reply information.

The highlighting and prompting module is used to select a target sub-template from at least one speech sub-template corresponding to the current business stage, and to highlight and prompt the target sub-template.

In some embodiments, the above-mentioned highlighting and prompting module includes:

The state obtaining unit is used to obtain the emotional state of the target user.

The second selection unit is configured to select a target sub-template from at least one speech sub-template corresponding to the current business stage according to the emotional state.

The embodiment of the present application discloses a call processing apparatus, which is used to obtain historical call records of a target user through a record acquisition module 201, where the historical call records include call records when the target user is on a call based on various types of speech templates; The feature extraction module 202 is configured to perform feature extraction respectively on the voice information and text information corresponding to the historical call records based on the preset bimodal neural network model, and obtain the voice features and text feature; the degree determination module 203 is used to determine the preference degree of the target user when facing various types of discourse templates according to the speech features and text features of the target user when facing various types of discourse templates; the prompt operation module 204 is used to use Based on the preference degree of the target user when facing various types of speech templates, the first target speech template with the highest preference is determined from the various types of speech templates, and the first target speech template is prompted to operate. Make the customer service agent communicate with the target user according to the first target speech template. In this way, the service quality of the telephone agent service is improved.

Correspondingly, an embodiment of the present application further provides a computer device, where the computer device may be a terminal. As shown in FIG. 3 , FIG. 3 is a schematic structural diagram of a computer device provided by an embodiment of the present application. The computer device 300 includes a processor 301 having one or more processing cores, a memory 302 having one or more computer-readable storage media, and a computer program stored on the memory 302 and executable on the processor. The processor 301 is electrically connected to the memory 302 . Those skilled in the art can understand that the computer device structure shown in the figures does not constitute a limitation on the computer device, and may include more or less components than the one shown, or combine some components, or arrange different components.

The processor 301 is the control center of the computer device 300, and uses various interfaces and lines to connect various parts of the entire computer device 300, by running or loading the software programs and/or modules stored in the memory 302, and calling the software programs and/or modules stored in the memory 302. to perform various functions of the computer device 300 and process data, so as to monitor the computer device 300 as a whole.

In this embodiment of the present application, the processor 301 in the computer device 300 loads the instructions corresponding to the processes of one or more application programs into the memory 302 according to the following steps, and the processor 301 executes the instructions stored in the memory. 302 application in order to achieve various functions:

Obtain the historical call records of the target user, the above-mentioned historical call records include the call records when the target user is called based on various types of speech templates;

Based on the preset dual-modal neural network model, feature extraction is performed on the voice information and text information corresponding to the historical call records, respectively, to obtain the voice features and text features of the target user when faced with various types of discourse templates;

Determine the preference degree of the target user when faced with various types of discourse templates according to the speech features and text characteristics of the target user when faced with various types of discourse templates;

Based on the preference degree of the target user when faced with various types of speech templates, the first target speech template with the highest preference is determined from the various types of speech templates, and the first target speech template is prompted to operate, so that the The customer service agent communicates with the target user according to the first target speech template.

For the specific implementation of the above operations, reference may be made to the foregoing embodiments, and details are not described herein again.

Optionally, as shown in FIG. 3 , the computer device 300 further includes: a touch display screen 303 , a radio frequency circuit 304 , an audio circuit 305 , an input unit 306 and a power supply 307 . The processor 301 is electrically connected to the touch display screen 303 , the radio frequency circuit 304 , the audio circuit 305 , the input unit 306 and the power supply 307 respectively. Those skilled in the art can understand that the computer device structure shown in FIG. 3 does not constitute a limitation to the computer device, and may include more or less components than the one shown, or combine some components, or arrange different components.

The touch screen 303 can be used to display a graphical user interface and receive operation instructions generated by the user acting on the graphical user interface. The touch display 303 may include a display panel and a touch panel. Among them, the display panel can be used to display messages input by the user or provided to the user and various graphical user interfaces of the computer equipment, which can be composed of graphics, text, icons, videos and any combination thereof. Optionally, the display panel may be configured in the form of a liquid crystal display (LCD, Liquid Crystal Displ y), an organic light emitting diode (OLED, Org nic Light-Emitting Diode). The touch panel can be used to collect the user's touch operations on or near it (such as the user's operations on or near the touch panel using a finger, stylus, etc., any suitable object or accessory), and generate corresponding operations instruction, and the operation instruction executes the corresponding program. Optionally, the touch panel may include two parts, a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch message from the touch detection device, converts it into contact coordinates, and then sends it to the touch controller. To the processor 301, and can receive the command sent by the processor 301 and execute it. The touch panel can cover the display panel, and when the touch panel detects a touch operation on or near it, it is transmitted to the processor 301 to determine the type of the touch event, and then the processor 301 provides the display panel according to the type of the touch event. Corresponding visual output. In this embodiment of the present application, the touch panel and the display panel may be integrated into the touch display screen 303 to implement input and output functions. However, in some embodiments, the touch panel and the touch panel may be used as two independent components to implement input and output functions. That is, the touch display screen 303 can also be used as a part of the input unit 306 to realize the input function. In this embodiment of the present application, the processor 301 executes the call processing program to display the speech template on the touch screen 303 .

The radio frequency circuit 304 can be used for transmitting and receiving radio frequency signals, so as to establish wireless communication with network equipment or other computer equipment through wireless communication, and send and receive signals with network equipment or other computer equipment.

Audio circuitry 305 may be used to provide an audio interface between the user and computer equipment through speakers, microphones. The audio circuit 305 can transmit the electrical signal converted from the received audio data to the speaker, and the speaker converts it into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is converted after being received by the audio circuit 305 In the form of audio data, the audio data is output to the processor 301 for processing, and then sent to, for example, another computer device via the radio frequency circuit 304, or the audio data is output to the memory 302 for further processing. Audio circuitry 305 may also include an ear jack to provide for communication of peripheral headphones with computer equipment.

Input unit 306 may be used to receive input numeric, character messages or user characteristic messages (eg fingerprint, iris, facial message, etc.), as well as generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control .

Power supply 307 is used to power various components of computer device 300 . Optionally, the power supply 307 may be logically connected to the processor 301 through a power management system, so that functions such as charging, discharging, and power consumption management are implemented through the power management system. The power source 307 may also include one or more DC or AC power sources, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and any other components.

Although not shown in FIG. 3 , the computer device 300 may further include a camera, a sensor, a Wi-Fi module, a Bluetooth module, and the like, which will not be repeated here.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

As can be seen from the above, the computer equipment provided in this embodiment obtains the historical call records of the target user, and the above-mentioned historical call records include the call records when the target user is called based on various types of speech templates; A neural network model is used to extract features from the voice information and text information corresponding to the historical call records, respectively, to obtain the voice features and text features of the target user when faced with various types of discourse templates; The speech features and text features of the template determine the preference degree of the target user when faced with various types of discourse templates; based on the target user's preference when faced with various types of discourse templates, the preferences are determined from various types of discourse templates The first target speech template with the highest degree is selected, and a prompt operation is performed on the first target speech template, so that the customer service agent can communicate with the target user according to the first target speech template.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above-mentioned embodiments can be completed by instructions, or by instructions that control relevant hardware, and the instructions can be stored in a computer-readable storage medium, and loaded and executed by the processor.

To this end, the embodiments of the present application provide a computer-readable storage medium, in which a plurality of computer programs are stored, and the computer programs can be loaded by a processor to execute any of the call processing methods provided by the embodiments of the present application. A step of. For example, the computer program may perform the following steps:

Obtain the historical call records of the target user, the above-mentioned historical call records include the call records when the target user is called based on various types of speech templates;

Based on the preset dual-modal neural network model, feature extraction is performed on the voice information and text information corresponding to the historical call records, respectively, to obtain the voice features and text features of the target user when faced with various types of discourse templates;

Determine the preference degree of the target user when faced with various types of discourse templates according to the speech features and text characteristics of the target user when faced with various types of discourse templates;

Based on the preference degree of the target user when faced with various types of speech templates, the first target speech template with the highest preference is determined from the various types of speech templates, and the first target speech template is prompted to operate, so that the The customer service agent communicates with the target user according to the first target speech template.

For the specific implementation of the above operations, reference may be made to the foregoing embodiments, and details are not described herein again.

Wherein, the storage medium may include: read-only memory (ROM, Re client account d Only Memory), random access memory (R client account M, R client account ndom client account access Memory), magnetic disk or optical disk Wait.

Since the computer program stored in the storage medium can execute the steps in any of the call processing methods provided by the embodiments of the present application, it is possible to realize the implementation of any call processing methods provided by the embodiments of the present application. For the beneficial effects, please refer to the previous embodiments for details, which will not be repeated here.

A call processing method, device, storage medium, and computer equipment provided by the embodiments of the present application have been described in detail above, and the principles and implementations of the present application are described with specific examples. The descriptions of the above embodiments are only It is used to help understand the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there will be changes in the specific embodiments and application scope. In summary, this specification The content should not be construed as a limitation on this application.

Claims (10)

1. A call processing method, characterized in that the method comprises: Obtaining the historical call records of the target user, the historical call records including the call records when the target user is on a call based on various types of discourse templates; Based on a preset dual-modal neural network model, feature extraction is performed on the voice information and text information corresponding to the historical call records, respectively, to obtain the voice features and text features of the target user when faced with various types of discourse templates; Determine the degree of preference of the target user when facing various types of discourse templates according to the speech features and text characteristics of the target user when facing various types of discourse templates; Based on the preference degree of the target user when facing various types of speech templates, a first target speech template with the highest preference degree is determined from the various types of speech templates, and the first target speech template A prompt operation is performed, so that the customer service agent communicates with the target user according to the first target speech template. 2. The method of claim 1, further comprising: When the target user is a new user, obtain attribute information of the target user; Determine historical users whose attributes are similar to the target user according to the attribute information of the target user; Obtain the second target speech template with the highest degree of preference corresponding to the historical user, and perform a prompt operation on the second target speech template, so that the customer service agent can match the second target speech template with the The target user makes a call. 3. The method of claim 1, further comprising: When the target user is a new user, acquiring call information within a preset time period of the target user; Determine the business scenario required by the target user according to the call information; Determine a third target speech template corresponding to the business scenario according to the business scenario, and perform a prompt operation on the third target speech template, so that the customer service agent can match the third target speech template with the third target speech template according to the third target speech template. the target user to make a call. 4. The method according to claim 1, wherein after the prompt operation is performed on the first target speech template, the method further comprises: Obtain the call records during the call in real time, input the call records into a preset neural network model to predict the preference degree, and obtain the preference degree of the target user for the first target speech template; When the preference degree is less than the preset threshold, select a fourth target speech template from the preset template set according to the call record, and perform a prompt operation on the fourth target speech template, so that the customer service agent Talk with the target user according to the fourth target speech template. 5. The method according to claim 4, wherein the selecting a fourth target speech template from a preset template set according to the call record comprises: Determine the business scenario currently required by the target user according to the call record; A fourth target speech template corresponding to the business scenario is selected from the template set. 6 . The method according to claim 1 , wherein the first target discourse template comprises at least one business phase, the business phase corresponds to at least one discourse sub-template, and the first target discourse template After the template performs the prompt operation, it also includes: Obtain the current reply information of the target user, and determine the current business stage according to the current reply information; A target sub-template is selected from at least one speech sub-template corresponding to the current business stage, and a highlighting prompt is given to the target sub-template. 7. The method according to claim 1, wherein the selecting a target sub-template from at least one speech sub-template corresponding to the current service stage comprises: obtaining the emotional state of the target user; A target sub-template is selected from at least one speech sub-template corresponding to the current business stage according to the emotional state. 8. A call processing device, wherein the device comprises: a record acquisition module, used for acquiring the historical call records of the target user, the historical call records including the call records when the target user is on a call based on various types of speech templates; The feature extraction module is used to perform feature extraction on the voice information and text information corresponding to the historical call records based on a preset dual-modal neural network model, respectively, to obtain the target user when faced with various types of discourse templates. Voice features and text features; a degree determination module, configured to determine the degree of preference of the target user when facing various types of discourse templates according to the speech features and text characteristics of the target user when facing various types of discourse templates; The prompt operation module is configured to, based on the preference degree of the target user when facing various types of speech templates, determine the first target speech template with the highest preference from the various types of speech templates, and assign the The first target speech template performs a prompting operation, so that the customer service agent can communicate with the target user according to the first target speech template. 9. A computer device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the program described in any one of claims 1 to 7 when the processor executes the program call handling method. 10 . A storage medium, characterized in that the storage medium stores a plurality of instructions, and the instructions are adapted to be loaded by a processor to execute the call processing method according to any one of claims 1 to 7 .

Images