Voice interaction method, device, electronic device and storage medium

Abstract

The present invention provides a voice interaction method, device, electronic device and storage medium, the method comprising: receiving a voice request, and determining a semantic classification result of the voice request; when the semantic classification result is within a preset corpus range, detecting the state of the target user information corresponding to the voice request, and obtaining state information; when the state information indicates that the target user information is empty, responding to the voice request, and sending an information acquisition request, the information acquisition request is used to obtain the target user information. The method, device, electronic device and storage medium provided by the present invention actively guide the user to set user information through voice dialogue, facilitate subsequent personalized interaction according to the target user information, thereby enhancing the naturalness and intimacy of the interaction, and, without the need for the user to set through the mobile phone app, not only improves the user experience, but also speeds up the response time of the system.

Description

Technical Field

[0001] The present invention relates to the field of electrical technology, and in particular to a voice interaction method, device, electronic equipment and storage medium. Background Art

[0002] With the advancement of Internet technology, electrical appliances with voice interaction functions are becoming more and more popular. In order to enhance the naturalness and intimacy of interaction, it is particularly important to set exclusive nicknames for users.

[0003] Currently, the way for users to set nicknames is usually to force users to manually enter nicknames on the mobile app (Application). For the use of the app, users need to have a process of understanding. This way of setting nicknames by users is not intelligent and has a poor user experience. Summary of the invention

[0004] The present invention provides a voice interaction method, device, electronic device and storage medium, which are used to solve the defect in the prior art that users manually set nicknames, resulting in poor user experience, and realize the setting of user information through voice dialogue.

[0005] The present invention provides a voice interaction method, comprising:

[0006] receiving a voice request and determining a semantic classification result of the voice request;

[0007] When the semantic classification result is within a preset corpus range, detecting a state of target user information corresponding to the voice request to obtain state information;

[0008] When the state information indicates that the target user information is empty, the voice request is responded to and an information acquisition request is sent, where the information acquisition request is used to acquire the target user information.

[0009] The present invention also provides a voice interaction device, comprising:

[0010] A receiving module, used for receiving a voice request and determining a semantic classification result of the voice request;

[0011] A detection module, configured to detect the state of the target user information corresponding to the voice request and obtain state information when the semantic classification result is within a preset corpus range;

[0012] The response module is used to respond to the voice request and send an information acquisition request when the state information indicates that the target user information is empty, wherein the information acquisition request is used to acquire the target user information.

[0013] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of any of the above-mentioned voice interaction methods are implemented.

[0014] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned voice interaction methods are implemented.

[0015] The voice interaction method, device, electronic device and storage medium provided by the present invention detect the state of the target user information corresponding to the voice request when it is determined that the semantic classification result of the voice request is within the preset corpus range, and when it is determined that the target user information is empty, respond to the voice request and send an information acquisition request, thereby actively guiding the user to set the user information through voice dialogue, facilitating subsequent personalized interaction according to the target user information, thereby enhancing the naturalness and intimacy of the interaction, and improving the trust between the user and the voice terminal product. In addition, there is no need for the user to set it through the mobile phone app, which not only improves the user experience but also speeds up the response time of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is one of the flow charts of the voice interaction method provided by the present invention;

[0018] Figure 2 This is the second flow chart of the voice interaction method provided by the present invention;

[0019] Figure 3 It is a structural schematic diagram of the voice interaction device provided by the present invention;

[0020] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Currently, the way users set nicknames is to force users to manually enter nicknames when registering voiceprints on the Zhijia app page on their mobile phones. For the use of the Zhijia app, users need to go through a process of understanding. This way of setting nicknames is not smart and has a poor user experience. In addition, since third-party service providers are involved, the voiceprint registration process itself is relatively complicated. This method of forcing users to manually enter nicknames undoubtedly increases the burden on the overall link and increases the system response time.

[0023] In this regard, an embodiment of the present invention provides a voice interaction method. Figure 1 is one of the flow charts of the voice interaction method provided by the present invention, such as Figure 1 As shown, the execution subject of the method is a processor, which can be installed on a voice terminal or a cloud server. The voice terminal here can be a smart speaker, a smart refrigerator, a smart air conditioner and other smart devices with voice interaction functions. The method includes:

[0024] Step 110, receiving a voice request and determining a semantic classification result of the voice request;

[0025] Step 120, when the semantic classification result is within the preset corpus range, detecting the state of the target user information corresponding to the voice request to obtain state information;

[0026] Step 130: When the state information indicates that the target user information is empty, respond to the voice request and send an information acquisition request, where the information acquisition request is used to acquire the target user information.

[0027] Specifically, considering that the voiceprint registration method that forces users to manually enter their nicknames on the mobile app side will lead to a poor user experience and increase the system response time, the embodiment of the present invention provides a method for actively guiding the target user to set information through voice dialogue. The target user only needs to have a voice dialogue with the voice interaction module installed on the voice terminal to complete the setting of the target user information, without the need to operate through the mobile app side or call a third-party interface, thereby not only improving the user experience but also speeding up the system response time. Here, the target user information may include the target user's nickname, age, occupation, interests and hobbies, etc., and the embodiment of the present invention does not specifically limit this.

[0028] Based on this, considering that when the user sends a voice request such as "shut down" or "lower the volume" to the voice terminal, it indicates that the user may want to end the voice conversation as soon as possible. If the voice guides the user to set information in this case, it will seriously affect the user's experience. In this regard, the voice interaction module in the embodiment of the present invention forwards the voice request sent by the target user to the processor after receiving it, and the processor performs semantic classification on the voice request to obtain the semantic classification result of the voice request. Then, the processor determines whether the semantic classification result is within the preset corpus range.

[0029] It should be noted that the preset corpus range may include multiple pre-set corpora, and the corpora here may be, for example, corpora that users frequently use, such as "I want to listen to a song", "What's the weather like today", etc., so as to ensure that the target user information is set as soon as possible, facilitate subsequent personalized interaction based on the target user information, and enhance the naturalness and intimacy of the interaction. Moreover, these corpora will not have the situation where the user wants to end the voice conversation as soon as possible, thereby avoiding the problem of subsequent voice guiding the target user to set information and resulting in poor user experience.

[0030] When the processor determines that the semantic classification result is within the preset corpus range, the processor can further detect whether the target user information corresponding to the voice request has been set, thereby obtaining status information. When the status information indicates that the target user information has not been set, that is, the target user information is empty, it means that the target user can be guided by voice to set the information: first, the processor can control the voice interaction module to respond to the voice request, and send an information acquisition request for obtaining the target user information, which can be achieved by issuing a voice for inquiring about the target user information, thereby achieving voice guidance for the target user to set the information.

[0031] For example, taking nickname setting as an example, the voice request sent by the user is "How is the weather today?", the voice interaction module of the voice terminal responds to the voice request, that is, replies to the user "Today's weather is 15-18 degrees Celsius, sunny", and sends an information acquisition request "Then Xiaoyou, can I ask you a question? How do you want me to call you?" On this basis, the processor can set the target user's nickname according to the voice of the user's reply received by the voice interaction module.

[0032] The method provided by the embodiment of the present invention detects the state of the target user information corresponding to the voice request when it is determined that the semantic classification result of the voice request is within the preset corpus range, and responds to the voice request and sends an information acquisition request when it is determined that the target user information is empty, thereby actively guiding the user to set the user information through voice dialogue, facilitating subsequent personalized interaction based on the target user information, thereby enhancing the naturalness and intimacy of the interaction, and improving the user's trust in the voice terminal product. In addition, there is no need for the user to set it through the mobile phone app, which not only improves the user experience but also speeds up the response time of the system.

[0033] Based on any of the above embodiments, in step 130, responding to the voice request and sending an information acquisition request includes:

[0034] Detecting the registration status of the voiceprint feature of the target user corresponding to the voice request;

[0035] When the registration status is registered, the voice request is answered and an information acquisition request is sent;

[0036] Otherwise, the voice request is answered.

[0037] Specifically, when it is determined that the target user information corresponding to the voice request is empty, the processor can further detect the registration status of the voiceprint feature of the target user corresponding to the voice request. The registration status here is used to indicate whether the voiceprint feature of the corresponding user has been registered. If the registration status is registered, it means that the target user can be guided by voice to set the information at this time, and then the voice interaction module can be controlled to respond to the voice request first, and then send an information acquisition request for obtaining the target user information. If the registration status is unregistered, it means that the user information setting cannot be completed at this time. The voice interaction module only needs to respond to the voice request, and does not need to send an information acquisition request for obtaining the target user information, thereby avoiding unnecessary voice interaction processes.

[0038] Based on any of the above embodiments, sending an information acquisition request further includes:

[0039] Receive the information reply voice corresponding to the information acquisition request;

[0040] Based on the information reply voice, the target user information is determined and bound to the voiceprint feature.

[0041] Specifically, after the voice interaction module issues an information acquisition request, the user will respond to the information acquisition request and send an information reply voice. On this basis, the processor can determine the target user information that the target user expects to set according to the information reply voice received by the voice interaction module, and bind the target user information with the voiceprint features of the target user who has previously completed registration, thereby facilitating the subsequent control of the voice interaction module to perform personalized interaction according to the target user information and enhance the naturalness and intimacy of the interaction.

[0042] Furthermore, the target user information can be obtained specifically in the following manner: after receiving the message reply voice of the user, the processor can perform voice recognition on the message reply voice, thereby obtaining the recognized text of the message reply voice, and then, considering that the recognized text includes not only text related to the user information but also other irrelevant texts, the processor can extract the information that the user expects to set from the recognized text, and determine the target user information based on the extracted information.

[0043] For example, in the above example, after the voice interaction module sends out an information acquisition request "Xiaoyou, can I ask you a question? What do you want me to call you?", the user responds to the information acquisition request, that is, replies to the voice interaction module of the voice terminal with an information reply voice. On this basis, the processor can perform voice recognition on the information reply voice received by the voice interaction module, obtain the recognized text "My name is Xiao Ming", and extract "Xiao Ming" from the recognized text, thereby determining the target user information, and binding the target user information with the corresponding voiceprint feature.

[0044] Based on any of the above embodiments, in step 130, responding to the voice request and sending an information acquisition request includes:

[0045] Find the information acquisition record corresponding to the voice request;

[0046] When the information acquisition record is empty, or the interval between the last acquisition time in the information acquisition record and the current time exceeds the preset time length, the voice request is responded to and the information acquisition request is sent;

[0047] Otherwise, the voice request is answered.

[0048] Specifically, in order to avoid frequent voice guidance of users to set information, resulting in a poor user experience, after determining that the target user information corresponding to the voice request is empty, the embodiment of the present invention can further search for the information acquisition record corresponding to the voice request. The information acquisition record here is a record made by the processor for sending an information acquisition request to the target user. When the information acquisition record is empty, it means that the processor has never guided the target user to set information by voice before, or when the interval between the last acquisition time in the information acquisition record and the current time exceeds the preset time, the last acquisition time is the time when the processor last sent an information acquisition request to the target user, indicating that the processor has not voice-guided the target user to set information within the preset time. In this case, the voice interaction module can be controlled to respond to the voice request first, and then the information acquisition request is sent to guide the target user to set information by voice.

[0049] Otherwise, that is, when the interval between the last acquisition time in the information acquisition record and the current time does not exceed the preset time length, it is only necessary to control the voice interaction module to respond to the voice request, and the entire process can be ended without issuing an information acquisition request. For example, the voice request is "How is the weather today?" and the preset time length is three months. If it is detected that the interval between the last acquisition time corresponding to the voice request and the current time is no more than three months, the voice interaction module can be controlled to reply "Weather ×××" to end the voice interaction process.

[0050] Here, the preset duration may be preset according to actual needs or may be set according to an empirical value. The unit of the preset duration may be day, month, hour, etc., which is not specifically limited in the embodiment of the present invention.

[0051] It should be noted that after determining that the target user information is empty, it does not mean that the processor has not sent an information acquisition request to the target user and set the target user information within the preset time, because there may be a situation where the target user information has been set but not successfully.

[0052] Based on any of the above embodiments, sending an information acquisition request further includes:

[0053] Control the target user information to be displayed in the display terminal;

[0054] In response to a modification operation of the target user based on the target user information, the target user information is updated.

[0055] Specifically, considering the diversity of Chinese characters with the same pronunciation, for example, the information reply voice of the user's reply is "Wo jiao xiao Ming", and the target user information determined by the processor according to the information reply voice is "Xiao Ming", but in fact the target user information that the user expects to set is "Xiao Ming", "Xiao Ming", etc., which causes a deviation in the setting of the target user information. In response to this problem, after sending an information acquisition request and acquiring the target user information, the embodiment of the present invention can also control the target user information to be displayed in the display terminal, so as to facilitate the user to view the target user information set by himself and modify the target user information. On this basis, when the processor receives the target user's modification operation on the target user information fed back by the display terminal, it can respond to the modification operation and update the target user information. Here, the display terminal can be a user app end or a display module on a voice terminal, and the embodiment of the present invention does not specifically limit this.

[0056] The method provided by the embodiment of the present invention controls the target user information to be displayed in the display terminal after the target user information is acquired, thereby facilitating the user to view the target user information set by the user, and providing the user with the right to modify and correct, so that the user can correct typos while viewing the target user information. In addition, the user can be provided with an entry to actively set the target user information, thereby facilitating the user to actively set the target user information he wants, and further improving the user experience.

[0057] Based on any of the above embodiments, detecting the state of the target user information corresponding to the voice request to obtain the state information, and then further comprising:

[0058] When the state information indicates that the target user information is not empty, the voice request is responded to based on the target user information.

[0059] Specifically, after obtaining the status information of the target user information corresponding to the voice request, if the status information indicates that the target user information is not empty, it can be determined that there is no need to guide the target user to set the information by voice, and the processor can control the voice interaction module to respond to the voice request according to the target user information. Here, the way to respond to the voice request can be to actively call the target user by the user nickname determined according to the target user information when responding to the voice request. For example, if the voice request is "How is the weather today?" and the target user information is "student", the voice interaction module can reply to the user "Hello, classmate, the weather today is ×××" when responding to the voice request. It can also be determined according to the target user information. The content of the response to the voice request can be played according to the target user information. Some audio content that is popular with the elderly can be played when responding to the voice request. The embodiment of the present invention does not specifically limit this.

[0060] The method provided by the embodiment of the present invention improves the personalization of voice interaction, strengthens the naturalness and intimacy of interaction, and enhances the trust between users and voice terminal products by responding to voice requests according to target user information.

[0061] Based on any of the above embodiments, responding to the voice request based on the target user information includes:

[0062] Determine the target user nickname based on the target user information;

[0063] Based on the target user nickname and the historical accuracy of the target user nickname, a target response language is determined, and the voice request is responded to based on the target response language, and the historical accuracy rate is the probability that the target user nickname matches the voice request.

[0064] Specifically, when the state information indicates that the target user information is not empty, the corresponding target user nickname can be determined according to the target user information. When responding to the voice request, the target user can be actively called according to the target user nickname, thereby enhancing the naturalness and intimacy of the interaction. It is understandable that if the target user information includes a user nickname, the user nickname included in the target user information can be directly used as the target user nickname. If the target user information does not include a user nickname, the corresponding target user nickname can be determined according to the occupation, age, gender and other information included in the target user information. For example, if the voice request is "How is the weather today?" and the target user information contains the occupation "student", the target user nickname can be "classmate". For example, if the target user information includes a user nickname and the historical accuracy of the target user nickname is low, the target user information can also be used according to the identity type corresponding to the target user information, such as children, adults, gender, elderly, etc., and the target response language is determined based on the identity type, and the voice request is responded to based on the target response language.

[0065] Based on this, if the voice terminal triggers the address "Hello, so-and-so" every time the user interacts with the voice terminal, the user will find this interaction annoying and the user experience is poor. To address this problem, after determining the target user nickname, the processor of the embodiment of the present invention can respond to the voice request according to the target user nickname and the historical accuracy rate. Here, the historical accuracy rate represents the probability of calling the target user by the target user nickname when responding to the voice request, that is, the matching probability of the target user nickname and the voice request.

[0066] The method provided by the embodiment of the present invention determines the target user nickname after setting the target user information, and triggers a response containing the target user nickname under a certain strategy, thereby avoiding frequent addressing of the target user and affecting the user experience.

[0067] Based on any of the above embodiments, when it is determined that the target user information corresponding to the voice request is empty, the processor can further detect the registration status of the voiceprint feature of the target user corresponding to the voice request. If the registration status is registered, it means that the target user has completed the voiceprint feature registration process, and the identity characteristics of the target user can be obtained from the context parameters obtained in the voiceprint feature registration process, and the corresponding identity title can be determined according to the identity characteristics. According to the identity title, the voice interaction module is controlled to respond to the voice request, thereby avoiding being too blunt when directly responding to the voice request and improving the user experience. Here, the identity characteristics can include the age, gender and other characteristics of the target user, and the identity title can be a general title determined according to the age, gender and other characteristics of the target user, such as handsome guy, beautiful girl, child, old man, little sister, little brother, etc. The embodiment of the present invention does not make specific limitations on this.

[0068] For example, the voice request is "How is the weather today?", and the identity characteristics of the target user corresponding to the voice request indicate that the target user's gender is male and the target user's age is 25 years old. Then the determined identity title can be "handsome guy". When responding to the voice request, the voice interaction module can reply to the target user "Hello, handsome guy, the weather today is ×××", and then send an information acquisition request "Then Xiaoyou, can I ask you a question? How do you want me to call you?"

[0069] Based on any of the above embodiments, an embodiment of the present invention provides a voice interaction method. Figure 2 This is the second flow chart of the voice interaction method provided by the present invention. Figure 2 As shown, taking the target user information as the user nickname as an example, the voice request sent by the target user is "How is the weather today?" After receiving the voice request, the voice interaction module forwards it to the processor. The processor first determines whether the voice request meets the trigger condition. The judgment method is to semantically classify the voice request, obtain the semantic classification result, and then determine whether the semantic classification result is within the preset corpus range: if the semantic classification result is not within the preset corpus range, the voice interaction module can be controlled to directly reply to the target user "Weather ×××". If the semantic classification result is within the preset corpus range, it can be further determined whether the target user information corresponding to the voice request is empty, that is, whether the user nickname is empty. If the user nickname is empty, the process of guiding the target user to set the nickname by voice can be entered. If the user nickname is not empty, it means that the user nickname has been set. At this time, the process of achieving the effect can be entered.

[0070] After entering the process of voice-guiding the target user to set a nickname, the processor will further determine whether the voiceprint feature of the target user has been registered. If not, the voice interaction module can be controlled to directly reply to the target user "Weather ×××"; if it has been registered, the corresponding information acquisition record of the target user can be further searched to determine whether an information acquisition request has been sent within the preset time to remind the target user to set a user nickname: if the information acquisition request has been sent within the preset time, the voice interaction module can be controlled to directly reply to the target user "Weather ×××"; if the information acquisition request has not been sent within the preset time, the voice interaction module can be controlled to reply to the target user "Weather ×××" and send an information acquisition request "Can Xiaoyou ask you a question? How do you want me to call you?" It should be noted that when sending an information acquisition request, in order to ensure the continuity of multiple rounds of dialogue, it is necessary to set multiple rounds of parameters and start multiple rounds of dialogue. In addition, while starting multiple rounds of dialogue, the use of the nickname guidance setting function can be recorded. Then, when the target user replies to the message with a voice reply "My name is so-and-so", the processor can reply to the voice reply according to the message received by the voice interaction module, set the user nickname, and control the voice interaction module to reply "OK, so-and-so", thereby ending the entire setting process.

[0071] After entering the process of achieving the effect, the processor can respond to the voice request based on the historical accuracy and user nickname. If it hits, the voice interaction module can be controlled to reply to the target user "Hello, XXX, weather ×××". If it does not hit, the voice interaction module can be controlled to directly reply to the target user "Weather ×××", thereby avoiding frequently calling the target user and affecting the user experience.

[0072] It should be noted that the order of the four judgment conditions, namely, whether the skill is triggered, whether the nickname is empty, whether the voiceprint feature has been registered, and whether the reminder has been given within the preset time, can be set according to needs. Figure 2 The sequence shown is only one implementation method, and the embodiment of the present invention does not specifically limit this.

[0073] The method provided in the embodiment of the present invention enables the user to guide the target user to set an exclusive nickname through voice dialogue when using the voice terminal, thereby enhancing the naturalness and intimacy of the interaction, making the interaction more humane, and improving the trust between the user and the voice terminal product. In addition, the user can query and modify his or her nickname through the smart home app on the mobile phone, thereby improving the user experience.

[0074] Based on any of the above embodiments, Redis is a high-performance key-value database that supports data persistence, multiple data structures, and data backup. Redis can reduce database pressure, improve overall performance, and bring a better user experience.

[0075] Based on this, the embodiment of the present invention uses Redis to cache historical accuracy and context parameters to facilitate the processor to read the historical accuracy and context parameters from the Redis cache, so that the entire user information setting process minimizes the use of the database and further improves the response time of the system. Here, the context parameters may include basic user information such as the user's identity characteristics, which may be obtained based on the voiceprint registration and recognition results, or may be configured. The embodiment of the present invention does not specifically limit this. In addition, Redis can dynamically configure the historical accuracy. After the processor determines that the semantic classification result is within the preset corpus range and the target user information corresponding to the voice request exists, it can quickly read the historical accuracy cached in Redis and respond to the voice request according to the dynamically configured historical accuracy.

[0076] In order to further improve the response time, the embodiment of the present invention also introduces websocket message middleware. Different from the existing message sending mechanism that needs to monitor whether the other party has received the message after sending it, the embodiment of the present invention uses the message middleware to send the message, and there is no need to pay attention to whether the other party has received it after sending it, thereby speeding up the response time of the system. Here, the message can be, for example, user information, scene coding and other information. In addition, the message middleware can also be used to record the use of the information guidance setting function, which is convenient for future data analysis and provides data support for future accurate recommendations.

[0077] Based on any of the above embodiments, Spring Boot is a new open source framework provided on the basis of Spring, which is designed to simplify the initial construction and development process of new Spring applications. The framework uses a specific method for configuration, so that developers no longer need to define boilerplate configurations. In this way, Spring Boot is committed to becoming a leader in the booming field of rapid application development.

[0078] Nacos (Dynamic Naming and Configuration Service) is a service infrastructure for building a modern application architecture centered on "services" (such as microservices paradigm, cloud native paradigm). It is dedicated to the discovery, management and information configuration of microservices, and can help developers quickly implement dynamic service discovery, service configuration, service metadata and traffic management, so as to build, deliver and manage microservice platforms more agilely and easily. Nacos supports the discovery, configuration and management of almost all mainstream types of services.

[0079] In this regard, the embodiment of the present invention uses Spring Boot to integrate Nacos as a dynamic configuration center to implement the setting of target user information. The dynamic configuration center can dynamically modify the configuration content and it can take effect without restarting.

[0080] The voice interaction device provided by the present invention is described below. The voice interaction device described below and the voice interaction method described above can be referenced to each other.

[0081] Based on any of the above embodiments, an embodiment of the present invention provides a voice interaction device. Figure 3 is a structural diagram of the voice interaction device provided by the present invention, such as Figure 3 As shown, the device comprises:

[0082] The receiving module 310 is used to receive a voice request and determine a semantic classification result of the voice request;

[0083] A detection module 320, configured to detect the state of the target user information corresponding to the voice request and obtain state information when the semantic classification result is within the preset corpus range;

[0084] The response module 330 is used to respond to the voice request and send an information acquisition request when the state information indicates that the target user information is empty, and the information acquisition request is used to acquire the target user information.

[0085] The device provided by the embodiment of the present invention detects the state of the target user information corresponding to the voice request when it is determined that the semantic classification result of the voice request is within the preset corpus range, and responds to the voice request and sends an information acquisition request when it is determined that the target user information is empty, thereby actively guiding the user to set the user information through voice dialogue, facilitating subsequent personalized interaction based on the target user information, thereby enhancing the naturalness and intimacy of the interaction, and improving the user's trust in the voice terminal product. In addition, there is no need for the user to set it through the mobile phone app, which not only improves the user experience but also speeds up the response time of the system.

[0086] Based on any of the above embodiments, responding to the voice request and sending an information acquisition request includes:

[0087] Detecting the registration status of the voiceprint feature of the target user corresponding to the voice request;

[0088] When the registration status is registered, the voice request is answered and an information acquisition request is sent;

[0089] Otherwise, the voice request is answered.

[0090] Based on any of the above embodiments, sending an information acquisition request further includes:

[0091] Receive the information reply voice corresponding to the information acquisition request;

[0092] Based on the information reply voice, the target user information is determined and bound to the voiceprint feature.

[0093] Based on any of the above embodiments, responding to the voice request and sending an information acquisition request includes:

[0094] Find the information acquisition record corresponding to the voice request;

[0095] When the information acquisition record is empty, or the interval between the last acquisition time in the information acquisition record and the current time exceeds the preset time length, the voice request is responded to and the information acquisition request is sent;

[0096] Otherwise, the voice request is answered.

[0097] Based on any of the above embodiments, sending an information acquisition request further includes:

[0098] Control the target user information to be displayed in the display terminal;

[0099] In response to a modification operation of the target user based on the target user information, the target user information is updated.

[0100] Based on any of the above embodiments, detecting the state of the target user information corresponding to the voice request to obtain the state information, and then further comprising:

[0101] When the state information indicates that the target user information is not empty, the voice request is responded to based on the target user information.

[0102] Based on any of the above embodiments, responding to the voice request based on the target user information includes:

[0103] Determine the target user nickname based on the target user information;

[0104] Based on the target user nickname and the historical accuracy of the target user nickname, a target response language is determined, and the voice request is responded to based on the target response language, and the historical accuracy rate is the probability that the target user nickname matches the voice request.

[0105] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute the voice interaction method, which includes: receiving a voice request and determining the semantic classification result of the voice request; when the semantic classification result is within the preset corpus range, detecting the state of the target user information corresponding to the voice request and obtaining the state information; when the state information indicates that the target user information is empty, responding to the voice request and sending an information acquisition request, the information acquisition request is used to acquire the target user information.

[0106] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0107] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the voice interaction method provided by the above methods, which includes: receiving a voice request and determining a semantic classification result of the voice request; when the semantic classification result is within a preset corpus range, detecting the status of the target user information corresponding to the voice request and obtaining status information; when the status information indicates that the target user information is empty, responding to the voice request and sending an information acquisition request, wherein the information acquisition request is used to obtain the target user information.

[0108] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the voice interaction method provided by the above-mentioned methods, the method comprising: receiving a voice request and determining a semantic classification result of the voice request; when the semantic classification result is within a preset corpus range, detecting the status of the target user information corresponding to the voice request and obtaining status information; when the status information indicates that the target user information is empty, responding to the voice request and sending an information acquisition request, the information acquisition request being used to obtain the target user information.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0111] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A voice interaction method, characterized in that: include: receiving a voice request and determining a semantic classification result of the voice request; When the semantic classification result is within a preset corpus range, detecting a state of target user information corresponding to the voice request to obtain state information; When the state information indicates that the target user information is empty, responding to the voice request and sending an information acquisition request, wherein the information acquisition request is used to acquire the target user information; When the state information indicates that the target user information is empty, the method further includes: Detecting a registration status of a voiceprint feature of a target user corresponding to the voice request; If the registration status is registered, the identity feature of the target user is obtained from the context parameter obtained in the voiceprint feature registration process, and a corresponding identity title is determined according to the identity feature, and the voice request is controlled to be answered according to the identity title; The step of responding to the voice request and sending an information acquisition request includes: Searching for an information acquisition record corresponding to the voice request; When the information acquisition record is empty, or the interval between the last acquisition time in the information acquisition record and the current time exceeds a preset time length, responding to the voice request and sending the information acquisition request; Otherwise, the voice request is answered.

2. The voice interaction method according to claim 1, characterized in that: The step of responding to the voice request and sending an information acquisition request includes: Detecting a registration status of a voiceprint feature of a target user corresponding to the voice request; When the registration status is registered, responding to the voice request and sending the information acquisition request; Otherwise, the voice request is answered.

3. The voice interaction method according to claim 2, characterized in that: The sending of the information acquisition request further includes: Receiving an information reply voice corresponding to the information acquisition request; Based on the information reply voice, the target user information is determined, and the target user information is bound to the voiceprint feature.

4. The voice interaction method according to any one of claims 1 to 3, characterized in that: The sending of the information acquisition request further includes: Controlling the target user information to be displayed on a display terminal; In response to a modification operation of the target user based on the target user information, the target user information is updated.

5. The voice interaction method according to any one of claims 1 to 3, characterized in that: The step of detecting the state of the target user information corresponding to the voice request to obtain the state information further includes: When the state information indicates that the target user information is not empty, the voice request is responded to based on the target user information.

6. The voice interaction method according to claim 5, characterized in that: The step of responding to the voice request based on the target user information includes: Based on the target user information, determine the target user nickname; Based on the target user nickname and the historical accuracy of the target user nickname, a target response language is determined, and the voice request is responded to based on the target response language, and the historical accuracy rate is the probability that the target user nickname matches the voice request.

7. A voice interaction device, characterized in that: include: A receiving module, used for receiving a voice request and determining a semantic classification result of the voice request; A detection module, configured to detect the state of the target user information corresponding to the voice request and obtain state information when the semantic classification result is within a preset corpus range; a response module, configured to respond to the voice request and send an information acquisition request when the status information indicates that the target user information is empty, wherein the information acquisition request is used to acquire the target user information; The response module is further used for: When the state information indicates that the target user information is empty, detecting a registration state of a voiceprint feature of the target user corresponding to the voice request; If the registration status is registered, the identity feature of the target user is obtained from the context parameter obtained in the voiceprint feature registration process, and a corresponding identity title is determined according to the identity feature, and the voice request is controlled to be answered according to the identity title; The response module is further used for: Searching for an information acquisition record corresponding to the voice request; When the information acquisition record is empty, or the interval between the last acquisition time in the information acquisition record and the current time exceeds a preset time length, responding to the voice request and sending the information acquisition request; Otherwise, the voice request is answered.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the voice interaction method as described in any one of claims 1 to 6 are implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the voice interaction method according to any one of claims 1 to 6 are implemented.

Images