An intelligent voice navigation method and device, a storage medium and an electronic device

Abstract

The application relates to an intelligent voice navigation method and device, a storage medium and electronic equipment, and relates to the technical field of voice navigation. The method comprises the following steps: acquiring an actual crowd type of a target user; determining an adaptation mode of voice navigation for the target user according to the actual crowd type; if the adaptation mode is a first mode, acquiring actual voice information of the target user, providing a matched first service to a terminal of the target user based on the actual voice information; and if the adaptation mode is a second mode, determining a second service required by the target user according to the actual crowd type, and providing the second service to the terminal of the target user. The application has the effect that the user can conveniently and directly reach the required service.

Description

Technical Field

[0001] This application relates to the field of voice navigation technology, specifically to an intelligent voice navigation method, device, storage medium, and electronic device. Background Technology

[0002] Interactive Voice Response (IVR) is an automated technology system widely used in telephone services. It allows users to interact with the system through voice or keypad input to obtain information or perform certain operations. This technology provides users with a convenient and efficient self-service experience, enabling them to complete information inquiries and business processing without human intervention, greatly improving service efficiency and user experience. Voice navigation is a key component of IVR; it guides users through the system and helps them select the desired services or information through voice prompts and instructions.

[0003] Currently, the common method for voice navigation for users is as follows: after a user dials the phone number of an existing IVR system, the system automatically answers and plays a pre-recorded welcome message and navigation menu options. Following the voice prompts, the user selects the corresponding navigation menu using buttons to proceed to the next level of the navigation menu, and so on, until they obtain the desired service. This method involves too many levels of voice navigation and requires user button operations, making the process cumbersome and preventing users from easily accessing the services they need. Summary of the Invention

[0004] To enable users to conveniently access the business services they need, this application provides an intelligent voice navigation method, device, storage medium, and electronic device.

[0005] The first aspect of this application provides an intelligent voice navigation method, specifically including:

[0006] The actual demographic type of the target users is identified, where the target users are those who consult and process business through voice navigation.

[0007] Based on the actual user group type, an adaptation mode for voice navigation for the target user is determined. The adaptation mode is either a first mode or a second mode. The first mode is used to identify the user's voice information and provide matching business services, while the second mode is used to recommend business services that are easy for the user to need.

[0008] If the adaptation mode is the first mode, then the actual voice information of the target user is obtained, and based on the actual voice information, a matching first service is provided to the target user's terminal.

[0009] If the adaptation mode is the second mode, then based on the actual population type, the second service required by the target user is determined, and the second service is provided to the target user's terminal.

[0010] By adopting the above technical solution, after obtaining the actual user demographics, a suitable navigation mode (i.e., an adaptation mode) is matched to the target user based on their demographic type during voice navigation. This allows the target user to more easily access the services they need. If the adaptation mode is the first mode, the target user's business needs are determined based on their actual voice information, and a first service tailored to those needs is provided. If the adaptation mode is the second mode, the business services that the target user is most likely to need are determined based on their demographic type, ultimately identifying the second service that the target user truly requires and providing this second service directly to the target user's device. This allows users to conveniently access the services they need.

[0011] Optionally, determining the adaptation mode for voice navigation for the target user based on the actual user group type specifically includes:

[0012] In the first mode, obtain the historical abnormal reasons for the abnormality in the voice information recognition of the first historical user, count the first occurrence number of each historical abnormal reason, and select a first number of historical abnormal reasons from each historical abnormal reason in descending order of the first occurrence number to determine the target abnormal reason;

[0013] When an identification anomaly occurs due to a single target anomaly, the corresponding historical user's historical population type is obtained. The second occurrence count of each historical population type is counted. The second number of historical population types is selected from each historical population type in descending order of the second occurrence count to determine the target population type. The target population type is the historical population type corresponding to a single target anomaly.

[0014] A first weight is determined for each of the target anomaly causes, and a second weight is determined for each of the target population types corresponding to each of the target anomaly causes. The first weight is the ratio of the first occurrence count of each target anomaly cause to the sum of the first occurrence counts of all target anomaly causes, and the second weight is the ratio of the second occurrence count of a single target population type corresponding to the target anomaly cause to the sum of the second occurrence counts of all corresponding target population types.

[0015] Based on the actual user group type, the first weight, and the corresponding second weights, an adaptation mode for voice navigation for the target user is determined.

[0016] By adopting the above technical solution, the greater the frequency of the first occurrence, the more likely the historical anomaly cause in the first mode is to lead to voice information recognition anomalies, thus identifying the target anomaly cause; the greater the frequency of the second occurrence, the more likely users of the corresponding historical user group type are to experience recognition anomalies due to the target anomaly cause, thus identifying the target user group type. Finally, by combining the first weight and the corresponding second weights, the probability of the target user experiencing voice information recognition anomalies due to each target anomaly cause, as well as the overall probability of voice information recognition anomalies, is determined. This allows for the targeted determination of a voice navigation mode suitable for the target user, thereby facilitating the target user's direct access to the required business services.

[0017] Optionally, determining the voice navigation adaptation mode for the target user based on the actual user group type, the first weight, and the corresponding second weights specifically includes:

[0018] If the actual population type exists in each target population type corresponding to the target anomaly cause, then the corresponding target anomaly cause will be identified as a key anomaly cause.

[0019] Calculate the first product of the first weight of each of the key abnormal causes and the second weight of the corresponding actual population type, and sum the first products to obtain the sum of the first products;

[0020] If the sum of the first products is greater than a preset first threshold, then the largest first product is selected from all the first products;

[0021] When the key anomaly cause corresponding to the maximum first product is a descriptive cause, the adaptation mode for voice navigation of the target user is determined to be the second mode; when the key anomaly cause corresponding to the maximum first product is not a descriptive cause, the adaptation mode for voice navigation of the target user is determined to be the first mode.

[0022] If the sum of the first products is not greater than a preset first threshold, then the adaptation mode for voice navigation for the target user is determined to be the first mode.

[0023] By adopting the above technical solution, the larger the first product, the more likely the target user is to experience recognition errors in the first mode due to corresponding key anomalies, resulting in the inability to match the target user with accurate business needs. If the sum of the first products is greater than a preset first threshold, it indicates that the overall probability of the target user experiencing recognition errors in the first mode is relatively high. When the key anomaly corresponding to the largest first product is a description-related reason, it indicates that the target user is more likely to experience voice information recognition errors due to description problems during voice input, thus determining that the first mode is not suitable for voice input, and therefore the second mode is determined as the appropriate adaptation mode. If the sum of the first products is not greater than the first threshold, it indicates that the overall probability of the target user experiencing recognition errors in the first mode is relatively low, thus the first mode is suitable for voice navigation. This provides a targeted adaptation mode when providing voice navigation for the target user, making it easier for the target user to directly access the required business services.

[0024] Optionally, determining the second business service needed by the target user based on the actual population type specifically includes:

[0025] Obtain the historical business services provided to the second historical user corresponding to the actual population type, count the first occurrence frequency of each historical business service, and select a third number of historical business services from each historical business service according to the first occurrence frequency to determine the target business service. The second historical user is the historical user who did not show any identification anomaly in the first mode.

[0026] Obtain historical keywords of each second historical user voice input corresponding to a single target service, count the second occurrence frequency of each historical keyword, and select a fourth number of historical keywords from each historical keyword in descending order of the second occurrence frequency to determine the target keyword, wherein the target keyword is the historical keyword corresponding to a single target service;

[0027] A third weight is determined for each of the target business services, and a fourth weight is determined for each of the target keywords corresponding to each target business service. The third weight is the ratio of the first occurrence frequency of each target business service to the sum of the first occurrence frequencies of all target business services. The fourth weight is the ratio of the second occurrence frequency of a single target keyword corresponding to a target business service to the sum of the second occurrence frequencies of all corresponding target keywords.

[0028] Based on the third weight and the corresponding fourth weights, the second business service required by the target user is determined.

[0029] By adopting the above technical solution, the higher the frequency of occurrence of a service, the more likely historical users are to require the corresponding historical service during voice navigation, thus identifying the target service. Furthermore, the higher the frequency of occurrence of a service, the more likely the corresponding historical keywords will appear in the voice information input by users requiring the target service, thus identifying the target keywords. Finally, by combining the third weight and the corresponding fourth weights, the target keywords that are likely to appear in the voice information corresponding to each target service are analyzed, thereby assisting in more accurately determining the business needs that match the target users.

[0030] Optionally, determining the second service required by the target user based on the third weight and the corresponding fourth weights specifically includes:

[0031] Calculate the second product of each of the third weights and the corresponding third weights, and select the largest second product from all the second products;

[0032] The target keyword corresponding to the largest second product is determined as the candidate keyword, and the recommendation order of the corresponding candidate keyword is determined according to the largest second product. The larger the largest second product, the higher the recommendation order of the corresponding candidate keyword.

[0033] According to the recommended order, the corresponding candidate keywords are pushed to the target user's terminal in sequence, and at least one final keyword is determined from the candidate keywords according to the first selection instruction sent by the target user's terminal.

[0034] If the final keyword exists among the target keywords corresponding to the target business service, then the corresponding target business service is identified as a key business service, and the third product of the third weight of each key business service and the fourth weight of each corresponding final keyword is calculated.

[0035] The sum of each of the third products is obtained to obtain the sum of the second products of the corresponding key services. If the sum of the second products is greater than a preset second threshold, the corresponding key service is pushed to the target user's terminal, and the second service required by the target user is determined according to the second selection instruction sent by the target user's terminal.

[0036] By adopting the above technical solution, the larger the maximum second product, the more likely the target user is to have a corresponding demand for the target business service. The corresponding candidate keywords will be recommended higher and pushed to the target user with priority, thereby quickly and accurately determining the target user's business needs.

[0037] Optionally, the method further includes:

[0038] Extract at least one actual keyword from the actual voice information. If at least one of the actual keywords exists among the target keywords corresponding to the target business service, then the corresponding target business service is identified as an important business service.

[0039] Calculate the fourth product of the third weight of each important business service and the fourth weight of each corresponding actual keyword, and sum the fourth products to obtain the sum of the corresponding third products;

[0040] If the sum of the third products is greater than the preset second threshold, then the corresponding important business service is identified as the verification business service. If the first business service is the verification business service, then the verification of the first business service is determined to be successful.

[0041] By employing the above technical solution, if the sum of the third products exceeds a preset second threshold, the target user is highly likely to require the corresponding important business service. Therefore, the corresponding important business service is identified as the verification business service. Finally, if the identified first business service is indeed the verification business service, it indicates that the first business service closely matches the target user's needs. The verification of the first business service is then confirmed as successful, thus ensuring the accuracy of the business service matching for the target user.

[0042] Optionally, after providing the matching first service to the target user's terminal, the method further includes:

[0043] When the service matching feedback information sent by the target user's terminal indicates a service mismatch, the verification service with the largest sum of the third products among the verification service services is selected as the final service service.

[0044] If the first business service is not the final business service, then the first business service is replaced with the final business service;

[0045] If the first service is the final service, then the first service is replaced by the remaining verification service with the largest sum of the third products among the remaining verification services, where the remaining verification service is a verification service other than the final service.

[0046] By adopting the above technical solution, if the service matching feedback information is that the service does not match, it means that the target user does not need this first service, and there may be an abnormality in the voice information recognition. In this case, if the first service is not the final service, the first service will be replaced with the final service. If the first service is the final service, the remaining verification service with the largest sum of the third products among the remaining verification services will be replaced with the first service, thereby achieving rapid and accurate adjustment of the service for the target user.

[0047] A second aspect of this application provides an intelligent voice navigation device, specifically comprising:

[0048] The information acquisition module is used to acquire the actual demographic type of the target users, which are users who conduct business inquiries and transactions through voice navigation.

[0049] The mode determination module is used to determine the adaptation mode for voice navigation of the target user based on the actual population type. The adaptation mode is a first mode or a second mode. The first mode is used to identify the user's voice information and provide matching business services, and the second mode is used to recommend business services that are easy for the user to need.

[0050] The first matching module is configured to, if the adaptation mode is the first mode, obtain the actual voice information of the target user, and provide a matching first service to the target user's terminal based on the actual voice information;

[0051] The second matching module is used to determine the second business service required by the target user based on the actual population type if the adaptation mode is the second mode, and to provide the second business service to the target user's terminal.

[0052] By adopting the above technical solution, the information acquisition module obtains the actual population type of the target user. Then, the mode determination module determines the appropriate adaptation mode for the target user based on the actual population type. Then, when the adaptation mode is the first mode, the first matching module provides the matching first service to the target user's terminal based on the actual voice information. Finally, when the adaptation mode is the second mode, the second matching module determines the second service required by the target user based on the actual population type.

[0053] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when loaded and executed by a processor, performs the steps of the method described in any one of the first aspects.

[0054] A fourth aspect of this application provides an electronic device, specifically comprising:

[0055] A processor, a memory, and a computer program stored in the memory and capable of running on the processor, the processor being configured to load and execute the computer program stored in the memory to cause the electronic device to perform the method as described in any one of the first aspects.

[0056] In summary, this application includes at least one of the following beneficial technical effects: After obtaining the actual demographic type of the target user, a suitable navigation mode for voice navigation is matched to the target user based on the actual demographic type, i.e., an adaptation mode, thereby facilitating the target user to directly access the required business services. If the adaptation mode is the first mode, the target user's business needs are determined based on the analysis of their actual voice information, and a first business service tailored to those needs is provided. If the adaptation mode is the second mode, the business service that the target user is most likely to need is determined based on the target user's actual demographic type, and finally, the second business service that the target user actually needs is determined and directly provided to the target user's terminal. This allows users to conveniently access the business services they need. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating an intelligent voice navigation method provided in an embodiment of this application;

[0058] Figure 2 This is a flowchart illustrating another intelligent voice navigation method provided in an embodiment of this application;

[0059] Figure 3 This is a schematic diagram of the structure of an intelligent voice navigation device provided in an embodiment of this application;

[0060] Figure 4 This is a schematic diagram of another intelligent voice navigation device provided in an embodiment of this application.

[0061] Explanation of reference numerals in the attached diagram: 11. Information acquisition module; 12. Pattern determination module; 13. First matching module; 14. Second matching module; 15. Business verification module; 16. Business adjustment module. Detailed Implementation

[0062] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0063] In the description of the embodiments in this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0064] In the description of the embodiments of this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, B existing alone, or A and B existing simultaneously. Furthermore, unless otherwise stated, the term "multiple" means two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.

[0065] See Figure 1 This application discloses a flowchart of an intelligent voice navigation method, which can be implemented using a computer program or run on an intelligent voice navigation device based on the von Neumann architecture. The computer program can be integrated into an application or run as a standalone utility application, specifically including:

[0066] S101: Obtain the actual demographic type of the target users.

[0067] Specifically, the target users are those who use voice navigation to inquire about and handle business matters. For example, a target user could be someone who is currently dialing the customer service hotline of a brand's product and using the voice navigation provided by the hotline to inquire about the product's after-sales service. The actual user group type can be understood as a group classification based on the target user's characteristics, behavior, and needs. Furthermore, the executing entity of the intelligent voice navigation method disclosed in this application can be the backend server of the target IVR system, specifically an independent physical server or a server cluster composed of multiple physical servers. The corresponding specific implementation scenario is: the target user dials the phone number of the target IVR system through a terminal, and the corresponding backend server provides matching voice navigation to the target user, enabling the target user to conveniently access the required business service. Business services can include transferring to human customer service, after-sales service inquiries, or bill inquiries. The terminal can be a smartphone or tablet.

[0068] Furthermore, during voice navigation for target users, one method to obtain the target user's actual demographic type is as follows: With permission granted, retrieve the target user's basic information stored on the platform associated with the target IVR system. For example, if the target IVR system is a customer service hotline for an online shopping platform, then the associated platform is that online shopping platform. Basic information includes, but is not limited to, gender, age, and occupation. This basic information is then input into a pre-defined type prediction model to obtain the target user's actual demographic type. The type prediction model can be a trained decision tree model or a trained recurrent neural network model. The training process involves training the model using basic information samples labeled with demographic types until the model converges.

[0069] S102: Determine the appropriate voice navigation mode for the target users based on the actual user demographics.

[0070] Specifically, in this embodiment, the voice navigation mode for the target user mainly includes a first mode and a second mode. The adaptation mode is either the first mode or the second mode. The first mode is used to identify the user's voice information and provide matching business services. The second mode is used to recommend business services that the user easily needs. For example, in the first mode, the target user only needs to input "Help me transfer to a human customer service representative" through voice input on the terminal. The backend server, based on Natural Language Processing (NLP) and a large language model, identifies and determines the business services needed by the target user and provides them directly, without having to go through multiple layers of cumbersome navigation menus and button combinations. In the second mode, the backend server analyzes and determines the business services that the target user may need and pushes them to the terminal for the target user to select. Here, the voice information refers to the information input by voice.

[0071] Further, a feasible method for determining the adaptation mode is as follows: Retrieve a preset record of historical anomaly causes, including: in the first mode, the reasons for anomalies recorded by staff when the target IVR system encountered anomalies in recognizing the voice input of each historical user. Based on the above historical anomaly cause records, obtain the historical anomaly causes for the anomalies in recognizing the voice information of the first historical user in the first mode, and count the first occurrence frequency of each historical anomaly cause. The higher the first occurrence frequency, the more likely the corresponding historical anomaly cause in the first mode is to lead to anomalies in voice information recognition. Here, anomalies refer to recognition errors or unsuccessful recognition. Historical anomaly causes include, but are not limited to, unclear expression of business requirements in the voice input (expression reasons), excessive noise, and excessively fast speech rate. Then, in descending order of the first occurrence frequency, select the first number of historical anomaly causes from the various historical anomaly causes to determine the target anomaly cause, that is, the anomaly cause that is likely to cause recognition anomalies.

[0072] Furthermore, when an identification anomaly occurs due to a single target anomaly, the corresponding historical user's historical population type is obtained. The second occurrence frequency of each historical population type is counted. The higher the second occurrence frequency, the more likely users of the corresponding historical population type are to have identification anomalies due to the target anomaly. Then, in descending order of the second occurrence frequency, the historical population types with the second highest frequency are selected from each historical population type to determine the target population type corresponding to the target anomaly, that is, the population type that is likely to have identification anomalies due to the target anomaly.

[0073] Furthermore, a first weight is determined for each target anomaly cause, which is the ratio of the first occurrence count of each target anomaly cause to the sum of the first occurrence counts of all target anomaly causes. A second weight is also determined for each target population type corresponding to the target anomaly cause, which is the ratio of the second occurrence count of a single target population type corresponding to the target anomaly cause to the sum of the second occurrence counts of all corresponding target population types.

[0074] After determining the first weight and the corresponding second weights, if the actual user type exists in the target user group types corresponding to the target anomaly cause, then the corresponding target anomaly cause is identified as a key anomaly cause. Next, the first product of the first weight of each key anomaly cause and the second weight of the corresponding actual user group type is calculated. The larger the first product, the more likely the target user is to experience recognition anomalies in the first mode due to the corresponding key anomaly cause, resulting in an inability to match the target user with accurate business needs. The sum of each first product is obtained; the larger the sum of the first products, the greater the overall probability of the target user experiencing recognition anomalies in the first mode. Furthermore, if the sum of the first products is greater than a preset first threshold, it indicates a high overall probability of recognition anomalies in the first mode. Then, the largest first product is selected from the various first products. If the key anomaly cause corresponding to the largest first product is a description-related reason, it means that the target user is more prone to voice information recognition anomalies due to description problems during voice input, making it impossible to match accurate business services. This indicates that the target user is not suitable for the first mode, and the second mode is determined as the adaptation mode for voice navigation for the target user. Conversely, if the key anomaly reason corresponding to the largest first product is not the stated reason, it means that the target user is not likely to have recognition anomalies due to statement problems during voice input. Instead, recognition anomalies may be caused by issues other than statement problems, such as low voice input volume. In this case, the first mode is determined to be the adaptation mode for voice navigation for the target user, and a reminder message for the key anomaly reason corresponding to the largest first product is sent to the terminal to reduce the risk of recognition anomalies.

[0075] Furthermore, if the sum of the first products is not greater than the first threshold, it indicates that the overall probability of the target user exhibiting recognition abnormalities in the first mode is relatively small. Therefore, it is suitable to use the first mode for voice navigation, and the first mode is determined as the adaptation mode for voice navigation of the target user.

[0076] S103: If the adaptation mode is the first mode, then obtain the actual voice information of the target user, and provide the matching first service to the target user's terminal based on the actual voice information.

[0077] Specifically, if the adaptation mode is the first mode, the actual voice information of the target user sent by the terminal is then recognized by Automatic Speech Recognition (ASR) technology and a large language model to determine the corresponding business requirements. Next, the corresponding first business service is matched from a pre-set business database, which includes various different business services. Finally, Text-to-Speech (TTS) technology is used to provide a corresponding voice response to the terminal, delivering the matched first business service to the target user's terminal.

[0078] S104: If the adaptation mode is the second mode, then determine the second business service required by the target user based on the actual user group type, and provide the second business service to the target user's terminal.

[0079] Specifically, if the adaptation mode is the second mode, it indicates that the target user is likely experiencing expression problems. To enable users to directly access the services they need, the historical voice navigation records are used to retrieve the services provided to users corresponding to the actual user type. The frequency of repetition of each service is counted; the higher the frequency, the more likely the user of the corresponding user type is to need the relevant service during voice navigation. At least one recommended service is selected from all services, provided that the frequency of repetition of the recommended service exceeds a preset threshold. These recommended services are then pushed to the target user's terminal via voice response. Finally, the selection command sent by the terminal is retrieved, and the recommended service corresponding to the selection command is identified as the second service, which is then provided to the target user's terminal. It should be noted that the historical voice navigation records include, but are not limited to, the user's user type, the input voice information, and the services provided to them.

[0080] See Figure 2This application discloses a flowchart of another intelligent voice navigation method, which can be implemented using a computer program or run on an intelligent voice navigation device based on the von Neumann architecture. The computer program can be integrated into an application or run as a standalone utility application, specifically including:

[0081] S201: Obtain the actual demographic type of the target users.

[0082] S202: Determine the appropriate voice navigation mode for the target users based on the actual user demographics.

[0083] S203: If the adaptation mode is the first mode, then obtain the actual voice information of the target user, and provide the matching first service to the target user's terminal based on the actual voice information.

[0084] For details, please refer to steps S101-S103, which will not be repeated here.

[0085] S204: If the adaptation mode is the second mode, then obtain the historical business services provided to the second historical users corresponding to the actual population type, count the first occurrence frequency of each historical business service, and select a third number of historical business services from each historical business service according to the first occurrence frequency to determine the target business service.

[0086] S205: Obtain the historical keywords of each second historical user voice input corresponding to a single target business service, count the second occurrence frequency of each historical keyword, and select the fourth number of historical keywords from each historical keyword in descending order of the second occurrence frequency to determine the target keyword.

[0087] Specifically, the second historical user refers to the historical user who did not exhibit recognition anomalies in the first mode, and the target keyword is the historical keyword corresponding to a single target service. If the adaptation mode is the second mode, then based on the actual user group type, the second service needed by the target user is determined. The specific process is as follows: Based on the aforementioned historical voice navigation records, the historical services provided to the second historical users corresponding to the actual user group type are obtained. The first occurrence frequency of each historical service is counted. The higher the first occurrence frequency, the more likely the second historical user is to need the corresponding historical service during voice navigation. Following the order of first occurrence frequency from highest to lowest, a third number of historical services are selected from each historical service to determine the target service, i.e., the service that users of the actual user group type are likely to need. Further, based on the aforementioned historical voice navigation records, the voice information input by each second historical user corresponding to a single target service is obtained, i.e., the voice information input by the second historical user who has already received the target service. Then, historical keywords are extracted from the voice information based on ASR technology. The second occurrence frequency of each historical keyword is counted. The higher the second occurrence frequency, the more likely the corresponding historical keyword will appear in the voice information input by the user who needs the target service. Next, in descending order of frequency of occurrence, the fourth number of historical keywords are selected from each historical keyword to determine the target keywords corresponding to the target business service, that is, keywords that are likely to appear in the input voice information.

[0088] S206: Determine the third weight of each target business service and the fourth weight of each target keyword corresponding to each target business service.

[0089] S207: Determine the second business service required by the target user based on the third weight and the corresponding fourth weights.

[0090] Specifically, in this embodiment, the third weight is the ratio of the first occurrence frequency of each target service to the sum of the first occurrence frequencies of all target services, and the fourth weight is the ratio of the second occurrence frequency of a single target keyword corresponding to a target service to the sum of the second occurrence frequencies of all corresponding target keywords. Next, the second product of each third weight and its corresponding components is calculated. The larger the second product, the more likely the target keyword is to appear in the voice information input by users of the actual user group when they need a single target service. The largest second product is selected from all the second products, and the target keyword corresponding to this largest second product is determined as the candidate keyword. That is, the keyword most likely to appear in the voice information input by users of the actual user group for each target service. Then, according to the largest second product, the recommendation order of the corresponding candidate keywords is determined. The larger the largest second product, the more likely the target user is to have a need for the corresponding target service, and the higher the recommendation of the corresponding candidate keyword, the more preferentially it is pushed to the target user, thereby quickly and accurately determining the target user's business needs. Further, according to the recommendation order, the corresponding candidate keywords are pushed to the target user's terminal sequentially, specifically in the form of voice response. Upon receiving the first selection instruction sent by the target user's terminal, at least one final keyword is determined from the candidate keywords, that is, a keyword that best matches the target user's business needs.

[0091] Furthermore, if the target keywords corresponding to the target business service contain final keywords, then the corresponding target business service is identified as a key business service. The third weight of each key business service and the third product of the fourth weight of each corresponding final keyword are calculated, and these third products are summed to obtain the sum of the second products of the corresponding key business services. The larger the sum of the second products, the greater the likelihood that the target user needs the corresponding key business service. If the sum of the second products is greater than a preset second threshold, it indicates a high probability that the target user needs the corresponding key business service. Therefore, the corresponding key business service (the business service that the target user is highly likely to need) is pushed to the target user's terminal for selection. Upon receiving the second selection instruction sent by the target user's terminal, the second business service needed by the target user is determined.

[0092] In other embodiments, when the adaptation mode is the first mode, at least one actual keyword is extracted from the actual voice information input by the target user. If at least one actual keyword exists among the target keywords corresponding to the target service, then the corresponding target service is identified as an important service. Next, the third weight of each important service and the fourth weight of each corresponding actual keyword are calculated as a fourth product. The sum of each fourth product is then obtained as the sum of the third products of the corresponding important services. The larger the sum of the third products, the greater the likelihood that the target user needs the corresponding important service, combined with the actual keywords appearing in the actual voice information. Further, if the sum of the third products is greater than a preset second threshold, the likelihood that the target user needs the corresponding important service is relatively high, and then the corresponding important service is identified as a verification service. Finally, if the identified first service is a verification service, it means that the first service is more in line with the needs of the target user, and the verification of the first service is determined to be successful, thereby ensuring the accuracy of the service matching for the target user.

[0093] In another embodiment, after providing the first service to the terminal, if the service matching feedback information sent by the terminal indicates a service mismatch, it means that the target user does not need this first service, and there may be an anomaly in voice information recognition. In this case, the verification service with the largest sum of its third products among the various verification services is selected and determined as the final service. If the first service is not the final service, it is replaced with the final service; if the first service is the final service, the remaining verification services with the largest sum of their third products among the remaining verification services replace the first service, thereby achieving rapid and accurate adjustment of the target user's service. The remaining verification services are those other than the final service.

[0094] The implementation principle of an intelligent voice navigation method according to an embodiment of this application is as follows: After obtaining the actual user demographic type, a suitable navigation mode, i.e., an adaptation mode, is matched for the target user based on the actual demographic type, thereby facilitating the target user to directly access the required business services. If the adaptation mode is the first mode, the target user's business needs are determined based on the actual voice information analysis, and a first business service that fits the business needs is provided. If the adaptation mode is the second mode, the business service that the target user is most likely to need is determined based on the actual demographic type of the target user, and finally, the second business service that the target user actually needs is determined and directly provided to the target user's terminal. This allows the user to conveniently access the required business services.

[0095] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0096] Please see Figure 3 This is a schematic diagram of the structure of the intelligent voice navigation device provided in the embodiments of this application. This device can be implemented through software, hardware, or a combination of both, becoming all or part of the device. The device includes an information acquisition module 11, a pattern determination module 12, a first matching module 13, and a second matching module 14.

[0097] Information acquisition module 11 is used to acquire the actual population type of the target users, who are users who consult and handle business through voice navigation.

[0098] The mode determination module 12 is used to determine the adaptation mode for voice navigation of the target user based on the actual population type. The adaptation mode is either the first mode or the second mode. The first mode is used to identify the user's voice information and provide matching business services, while the second mode is used to recommend business services that are easy for the user to need.

[0099] The first matching module 13 is used to obtain the actual voice information of the target user if the adaptation mode is the first mode, and provide the matching first business service to the target user's terminal based on the actual voice information.

[0100] The second matching module 14 is used to determine the second business service required by the target user based on the actual population type if the adaptation mode is the second mode, and to provide the second business service to the target user's terminal.

[0101] Optionally, the pattern determination module 12 is specifically used for:

[0102] In the first mode, obtain the historical reasons for the abnormality in the voice information recognition of the first historical user, count the first occurrence of each historical reason, and select the first number of historical reasons for the target abnormality in descending order of the first occurrence.

[0103] When an identification anomaly occurs due to a single target anomaly, the corresponding historical user's historical population type is obtained. The second occurrence count of each historical population type is counted. Based on the second occurrence count in descending order, the second number of historical population types is selected from each historical population type to determine the target population type. The target population type is the historical population type corresponding to the single target anomaly.

[0104] Determine the first weight for each target anomaly cause and the second weight for each target population type corresponding to each target anomaly cause. The first weight is the ratio of the first occurrence of each target anomaly cause to the sum of the first occurrences of all target anomalies. The second weight is the ratio of the second occurrence of a single target population type corresponding to the target anomaly cause to the sum of the second occurrences of all target population types.

[0105] Based on the actual user group type, the first weight, and the corresponding second weights, determine the appropriate voice navigation mode for the target users.

[0106] Optionally, the pattern determination module 12 is specifically used for:

[0107] If there is an actual population type among the target population types corresponding to the target anomaly cause, then the corresponding target anomaly cause will be identified as a key anomaly cause.

[0108] Calculate the first product of the first weight of each key abnormal cause and the second weight of the corresponding actual population type, and sum the first products to obtain the sum of the first products;

[0109] If the sum of the first products is greater than a preset first threshold, then the largest first product is selected from all the first products;

[0110] When the key anomaly cause corresponding to the largest first product is the stated cause, the adaptation mode for voice navigation of the target user is determined to be the second mode; when the key anomaly cause corresponding to the largest first product is not the stated cause, the adaptation mode for voice navigation of the target user is determined to be the first mode.

[0111] If the sum of the first products is not greater than the preset first threshold, then the first mode is determined as the adaptation mode for voice navigation for the target user.

[0112] Optionally, the second matching module 14 is specifically used for:

[0113] Obtain the historical business services provided to the second historical users corresponding to the actual population type, count the first occurrence frequency of each historical business service, and select a third number of historical business services from each historical business service according to the first occurrence frequency to determine the target business service. The second historical users are the historical users who did not show any identification anomalies in the first mode.

[0114] Obtain the historical keywords of each second historical user voice input corresponding to a single target business service, count the second occurrence frequency of each historical keyword, and select the fourth number of historical keywords from each historical keyword in descending order of the second occurrence frequency to determine the target keyword. The target keyword is the historical keyword corresponding to a single target business service.

[0115] The third weight of each target business service is determined, and the fourth weight of each target keyword corresponding to each target business service is determined. The third weight is the ratio of the first occurrence frequency of each target business service to the sum of the first occurrence frequencies of all target business services. The fourth weight is the ratio of the second occurrence frequency of a single target keyword corresponding to a target business service to the sum of the second occurrence frequencies of all corresponding target keywords.

[0116] Based on the third weight and the corresponding fourth weights, determine the second business service required by the target user.

[0117] Optionally, the second matching module 14 is specifically used for:

[0118] Calculate the second product of each third weight and the corresponding third weight, and select the largest second product from all the second products;

[0119] The target keyword corresponding to the largest second product is determined as the candidate keyword, and the recommendation order of the candidate keyword is determined according to the largest second product. The larger the largest second product, the higher the recommendation order of the candidate keyword.

[0120] According to the recommendation order, the corresponding candidate keywords are pushed to the target user's terminal in sequence, and at least one final keyword is determined from the candidate keywords according to the first selection instruction sent by the target user's terminal.

[0121] If there is a final keyword among the target keywords corresponding to the target business service, then the corresponding target business service is identified as a key business service, and the third product of the third weight of each key business service and the fourth weight of each corresponding final keyword is calculated.

[0122] Summing up each third product yields the sum of the second products of the corresponding key services. If the sum of the second products is greater than a preset second threshold, the corresponding key service is pushed to the target user's terminal. Based on the second selection instruction sent by the target user's terminal, the second service required by the target user is determined.

[0123] Optional, such as Figure 4 As shown, the device also includes a service verification module 15, specifically used for:

[0124] Extract at least one actual keyword from the actual voice information. If at least one actual keyword exists among the target keywords corresponding to the target business service, then the corresponding target business service is identified as an important business service.

[0125] Calculate the fourth product of the third weight of each important business service and the fourth weight of each corresponding actual keyword, and sum the fourth products to obtain the sum of the corresponding third products;

[0126] If the sum of the third product is greater than the preset second threshold, then the corresponding important business service is identified as the verification business service. If the first business service is the verification business service, then the verification of the first business service is determined to be successful.

[0127] Optionally, the device also includes a service adjustment module 16, specifically used for:

[0128] When the service matching feedback information sent from the target user's terminal indicates a service mismatch, the service with the largest sum of the third products among the verified service services is selected as the final service.

[0129] If the first business service is not the final business service, then the first business service will be replaced with the final business service.

[0130] If the first service is the final service, then the first service is replaced by the remaining verification service with the largest sum of the third products among all remaining verification services. The remaining verification services are verification services other than the final service.

[0131] It should be noted that the intelligent voice navigation device provided in the above embodiments is only illustrated by the division of the above functional modules when executing the intelligent voice navigation method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent voice navigation device and the intelligent voice navigation method embodiment provided in the above embodiments belong to the same concept, and the implementation process is detailed in the method embodiment, which will not be repeated here.

[0132] This application also discloses a computer-readable storage medium, which stores a computer program, wherein when the computer program is executed by a processor, it employs an intelligent voice navigation method as described in the above embodiments.

[0133] The computer program can be stored in a computer-readable medium. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or certain middleware. The computer-readable medium includes any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the computer-readable medium includes, but is not limited to, the above-mentioned components.

[0134] The intelligent voice navigation method of the above embodiment is stored in the computer-readable storage medium and loaded and executed on the processor to facilitate the storage and application of the above method.

[0135] This application also discloses an electronic device in which a computer program is stored in a computer-readable storage medium. When the computer program is loaded and executed by a processor, it employs the aforementioned intelligent voice navigation method.

[0136] The electronic device can be a desktop computer, a laptop computer, or a cloud server, and includes, but is not limited to, a processor and a memory. For example, the electronic device may also include input / output devices, network access devices, and buses.

[0137] The processor can be a central processing unit (CPU). Of course, depending on the actual use, it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc., and this application does not limit it.

[0138] The memory can be an internal storage unit of an electronic device, such as a hard disk or RAM, or an external storage device, such as a plug-in hard disk, smart memory card (SMC), secure digital card (SD), or flash memory card (FC) equipped on the electronic device. Furthermore, the memory can be a combination of an internal storage unit and an external storage device. The memory is used to store computer programs and other programs and data required by the electronic device. The memory can also be used to temporarily store data that has been output or will be output. This application does not limit this.

[0139] In this electronic device, the intelligent voice navigation method of the above embodiment is stored in the memory of the electronic device and loaded and executed on the processor of the electronic device for convenient use.

[0140] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. An intelligent voice navigation method, characterized in that, The method includes: The actual demographic type of the target users is identified, where the target users are those who consult and process business through voice navigation. Based on the actual user group type, an adaptation mode for voice navigation for the target user is determined. The adaptation mode is either a first mode or a second mode. The first mode is used to identify the user's voice information and provide matching business services, while the second mode is used to recommend business services that are easy for the user to need. If the adaptation mode is the first mode, then the actual voice information of the target user is obtained, and based on the actual voice information, a matching first service is provided to the target user's terminal. If the adaptation mode is the second mode, then based on the actual population type, determine the second business service required by the target user, and provide the second business service to the target user's terminal; Based on the actual user demographics, determine the appropriate voice navigation mode for the target user, including: obtaining historical anomaly reasons for abnormal voice information recognition of a first historical user under the first mode; counting the first occurrence count of each historical anomaly reason; and selecting a first number of historical anomaly reasons as target anomaly reasons in descending order of the first occurrence count; obtaining the historical user demographics corresponding to the first historical user when an abnormality occurs due to a single target anomaly reason; counting the second occurrence count of each historical user demographics; and selecting a second number of historical anomaly reasons as target anomaly reasons in descending order of the second occurrence count. The historical user group type is determined as the target user group type, where the target user group type is the historical user group type corresponding to a single target anomaly cause. A first weight is determined for each target anomaly cause, and a second weight is determined for each target user group type corresponding to each target anomaly cause. The first weight is the ratio of the first occurrence count of each target anomaly cause to the sum of the first occurrence counts of all target anomaly causes, and the second weight is the ratio of the second occurrence count of a single target user group type corresponding to a target anomaly cause to the sum of the second occurrence counts of all corresponding target user groups. Based on the actual user group type, the first weight, and the corresponding second weights, an adaptation mode for voice navigation for the target user is determined.

2. The intelligent voice navigation method according to claim 1, characterized in that, The step of determining the voice navigation adaptation mode for the target user based on the actual user type, the first weight, and the corresponding second weights specifically includes: If the actual population type exists in each target population type corresponding to the target anomaly cause, then the corresponding target anomaly cause will be identified as a key anomaly cause. Calculate the first product of the first weight of each of the key abnormal causes and the second weight of the corresponding actual population type, and sum the first products to obtain the sum of the first products; If the sum of the first products is greater than a preset first threshold, then the largest first product is selected from all the first products; When the key anomaly cause corresponding to the maximum first product is a descriptive cause, the adaptation mode for voice navigation of the target user is determined to be the second mode; when the key anomaly cause corresponding to the maximum first product is not a descriptive cause, the adaptation mode for voice navigation of the target user is determined to be the first mode. If the sum of the first products is not greater than a preset first threshold, then the adaptation mode for voice navigation for the target user is determined to be the first mode.

3. The intelligent voice navigation method according to claim 1, characterized in that, The step of determining the second business service needed by the target user based on the actual population type specifically includes: Obtain the historical business services provided to the second historical user corresponding to the actual population type, count the first occurrence frequency of each historical business service, and select a third number of historical business services from each historical business service according to the first occurrence frequency to determine the target business service. The second historical user is the historical user who did not show any identification anomaly in the first mode. Obtain historical keywords of each second historical user voice input corresponding to a single target service, count the second occurrence frequency of each historical keyword, and select a fourth number of historical keywords from each historical keyword in descending order of the second occurrence frequency to determine the target keyword, wherein the target keyword is the historical keyword corresponding to a single target service; A third weight is determined for each of the target business services, and a fourth weight is determined for each of the target keywords corresponding to each target business service. The third weight is the ratio of the first occurrence frequency of each target business service to the sum of the first occurrence frequencies of all target business services. The fourth weight is the ratio of the second occurrence frequency of a single target keyword corresponding to a target business service to the sum of the second occurrence frequencies of all corresponding target keywords. Based on the third weight and the corresponding fourth weights, the second business service required by the target user is determined.

4. The intelligent voice navigation method according to claim 3, characterized in that, The step of determining the second service required by the target user based on the third weight and the corresponding fourth weights specifically includes: Calculate the second product of each of the third weights and the corresponding third weights, and select the largest second product from all the second products; The target keyword corresponding to the largest second product is determined as the candidate keyword, and the recommendation order of the corresponding candidate keyword is determined according to the largest second product. The larger the largest second product, the higher the recommendation order of the corresponding candidate keyword. According to the recommended order, the corresponding candidate keywords are pushed to the target user's terminal in sequence, and at least one final keyword is determined from the candidate keywords according to the first selection instruction sent by the target user's terminal. If the final keyword exists among the target keywords corresponding to the target business service, then the corresponding target business service is identified as a key business service, and the third product of the third weight of each key business service and the fourth weight of each corresponding final keyword is calculated. The sum of each of the third products is obtained to obtain the sum of the second products of the corresponding key services. If the sum of the second products is greater than a preset second threshold, the corresponding key service is pushed to the target user's terminal, and the second service required by the target user is determined according to the second selection instruction sent by the target user's terminal.

5. The intelligent voice navigation method according to claim 3, characterized in that, The method further includes: Extract at least one actual keyword from the actual voice information. If at least one of the actual keywords exists among the target keywords corresponding to the target business service, then the corresponding target business service is identified as an important business service. Calculate the fourth product of the third weight of each important business service and the fourth weight of each corresponding actual keyword, and sum the fourth products to obtain the sum of the corresponding third products; If the sum of the third products is greater than the preset second threshold, then the corresponding important business service is identified as the verification business service. If the first business service is the verification business service, then the verification of the first business service is determined to be successful.

6. The intelligent voice navigation method according to claim 5, characterized in that, After providing the matching first service to the target user's terminal, the method further includes: When the service matching feedback information sent by the target user's terminal indicates a service mismatch, the verification service with the largest sum of the third products among the verification service services is selected as the final service service. If the first business service is not the final business service, then the first business service is replaced with the final business service; If the first service is the final service, then the first service is replaced by the remaining verification service with the largest sum of the third products among the remaining verification services, where the remaining verification service is a verification service other than the final service.

7. An intelligent voice navigation device, used to implement the intelligent voice navigation method according to any one of claims 1 to 6, characterized in that, include: The information acquisition module (11) is used to acquire the actual population type of the target user, wherein the target user is a user who conducts business consultation and processing through voice navigation; The mode determination module (12) is used to determine the adaptation mode for voice navigation of the target user according to the actual population type. The adaptation mode is either a first mode or a second mode. The first mode is used to identify the user's voice information and provide matching business services. The second mode is used to recommend business services that are easy for the user to need. The first matching module (13) is used to obtain the actual voice information of the target user if the adaptation mode is the first mode, and provide the target user's terminal with a matching first service based on the actual voice information. The second matching module (14) is used to determine the second business service required by the target user based on the actual population type if the adaptation mode is the second mode, and to provide the second business service to the target user's terminal.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is loaded and executed by the processor, it employs the method described in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, When the processor loads and executes the computer program, it employs the method described in any one of claims 1-6.

Images