Method, device and equipment for processing vehicle-machine voice interaction data

Abstract

The present application provides a method, device and equipment for processing vehicle-computer voice interaction data. The method obtains voice log data including historical response voices of successful interactions, historical recognition operations corresponding to each historical response voice, and historical response services corresponding to each historical response voice, and then performs quantitative statistics on each historical response voice, each historical recognition operation and at least one of each historical response service, to obtain at least one of the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services, thereby realizing statistics on vehicle-computer voices of successful interactions. The statistical results can be used for function development optimization, solving the problem of high manual research costs, and the development optimization direction determined by the statistical results of the successful interaction voices has high accuracy.

Description

Technical Field

[0001] The present application relates to the field of voice processing technology, and in particular to a method, device and equipment for processing vehicle-machine voice interaction data. Background Art

[0002] With the development of automotive electronics, more and more car companies are choosing to install intelligent voice interaction systems in their vehicles. Among them, intelligent voice interaction systems can be used to allow users to interact with smart cars through voice commands. For example, users can use voice commands to call car settings, car control, map navigation, music and multimedia services, or check the weather, chat, etc.

[0003] In the process of developing intelligent voice interaction systems, technicians usually optimize or develop functions based on market research. However, manual research consumes a lot of manpower and time costs, and the accuracy of the research results is low, which may lead to the wrong direction of optimization and development, thus affecting the user experience. Summary of the invention

[0004] In view of this, the purpose of this application is to propose a method, device and equipment for processing vehicle-machine voice interaction data to solve the technical problems of high cost of manual research and wrong optimization development direction in the prior art.

[0005] Based on the above purpose, the present application provides a method for processing vehicle-machine voice interaction data, including:

[0006] Acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice;

[0007] Performing quantitative statistics on at least one of each of the historical response voices, each of the historical recognition operations, and each of the historical response services in the voice log data to obtain an interaction number statistical result;

[0008] The interaction number statistics include at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services.

[0009] Based on the above purpose, the present application also provides a vehicle-machine voice interaction data processing device, including:

[0010] A log acquisition module, used to acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice;

[0011] A log statistics module, used to perform quantitative statistics on at least one of the historical response voices, the historical recognition operations, and the historical response services in the voice log data to obtain an interaction number statistical result;

[0012] The interaction number statistics include at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services.

[0013] Based on the above purpose, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements a method for processing vehicle-machine voice interaction data as provided in any embodiment of the present application.

[0014] Based on the above purpose, the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the method for processing vehicle-machine voice interaction data provided in any embodiment of the present application.

[0015] From the above, it can be seen that the method for processing vehicle-machine voice interaction data provided by the present application obtains voice log data including each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice, and then performs quantitative statistics on each historical response voice, each historical recognition operation, and at least one of each historical response service, to obtain at least one of the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services, thereby realizing statistics on each vehicle-machine voice of successful interaction. The statistical results can be used for function development or function optimization without manual research, thus solving the problem of manual research consuming a lot of costs. In addition, the statistical results of successful interactions determined based on historical voice interaction data can reflect the actual usage of users. The development direction or optimization direction determined by the statistical results is highly accurate, thus avoiding the situation where the optimization development direction is wrong. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flowchart of a method for processing vehicle-machine voice interaction data provided in an embodiment of the present application;

[0018] Figure 2 A flowchart of another method for processing vehicle-machine voice interaction data provided in an embodiment of the present application;

[0019] Figure 3 A process of processing vehicle-machine voice interaction data provided by an embodiment of the present application is shown in FIG.

[0020] Figure 4 The structure and operation of a vehicle-machine voice interaction data processing device provided in an embodiment of the present application are shown;

[0021] Figure 5 The structure and operation of an electronic device provided in an embodiment of the present application are shown. DETAILED DESCRIPTION

[0022] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0023] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0024] Figure 1The flowchart of a method for processing vehicle-machine voice interaction data provided by an embodiment of the present application. The method can be executed by a vehicle-machine voice interaction data processing device, which can be implemented in software and / or hardware, and can be configured in an electronic device. Figure 1 As shown, the method may specifically include the following steps:

[0025] S110. Acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice.

[0026] The voice log data may be log data related to the car machine voice that has been served, that is, log data related to the car machine voice that has successfully interacted. It may also be log data related to the car machine voice that has successfully interacted within a preset period of time (for example, a few months or weeks). Specifically, the voice log data includes each historical response voice of successful interaction, the historical recognition operations corresponding to each historical response voice, and the historical response services corresponding to each historical response voice.

[0027] In this embodiment, the historical response voice may be a historical voice initiated by the user and responded to by the vehicle computer. Exemplarily, the voice log data may include multiple "topic": "dm.output", wherein each "dm.output" contains "input": "xxxxx", specifically, "xxxxx" is the historical response voice.

[0028] The historical recognition operation corresponding to the historical response voice can be the operation intention recognized for the historical response voice, such as planning a route, making a call or navigation. For example, the historical response voice is "How many degrees is it today?" or "Will it rain soon?", and the corresponding historical recognition operation is "Check the weather"; the historical response voice is "Change a song" or "Don't play it yet", and the corresponding historical recognition operation is "Playback control". Exemplarily, each "dm.output" also contains "intentName":"xxxxx", specifically, "xxxxx" is the historical recognition operation.

[0029] The historical response service corresponding to the historical response voice can be a skill triggered by the historical recognition corresponding to the historical response voice; such as weather, navigation, vehicle control and settings, jokes or poems, etc. For example, the historical response voice is "How long does it take to get to the company", and the corresponding historical response service is the in-vehicle general navigation. Exemplarily, each "dm.output" also includes a skill field, and the content in the skill field is the historical response service. In the process of the car computer providing voice services to users, the car computer can recognize the corresponding operation according to the voice issued by the user, and then determine the response service according to the operation.

[0030] Exemplarily, the voice log data can be obtained through a pre-established voice access layer. In a specific implementation, the voice log data of the user can be collected by the voice server and stored in an open source stream processing platform (such as Kafka); further, a distributed file system (such as Hadoop), a data warehouse tool (such as HIVE) or a scheduler can read the voice log data from Kafka to further analyze the voice log data through a relational database service (RDS) or a relational database management system (such as MySQL).

[0031] It should be noted that, considering that the voice log data of the test environment and the voice log data of the formal environment are stored in the voice server, the voice log data of the test environment may be the voice log data generated on the test device, and the voice log data of the formal environment may be the voice log data generated on the actual vehicle computer. In order to avoid the influence of the voice log data of the test environment on the statistical results of the number of interactions, all voice log data can be screened to extract the voice log data of the formal environment. Exemplarily, after reading all the voice log data from Kafka, the voice log data of the test environment can be eliminated through the device identifier of the test device. In this way, the influence of the voice log data generated on the test device on the statistical results can be avoided, so that the statistical results are more in line with the actual user usage, and the accuracy of the development direction or optimization direction is further improved.

[0032] S120, performing quantitative statistics on at least one of each historical response voice, each historical recognition operation, and each historical response service in the voice log data to obtain an interaction number statistical result.

[0033] Among them, the interaction number statistics result includes at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services.

[0034] Specifically, when an interactive statistical instruction is detected, a target statistical item corresponding to the interactive statistical instruction may be determined; wherein the target statistical item may be a voice item, an operation item or a service item, or any combination of the above three situations.

[0035] In this embodiment, if the target statistical item corresponding to the interactive statistical instruction includes a voice item, it means that it is necessary to perform quantitative statistics on each historical response voice in the voice log data; if the target statistical item corresponding to the interactive statistical instruction includes an operation item, it means that it is necessary to perform quantitative statistics on each historical recognition operation in the voice log data; if the target statistical item corresponding to the interactive statistical instruction includes a service item, it means that it is necessary to perform quantitative statistics on each historical response service in the voice log data.

[0036] Furthermore, according to the target statistical item, at least one of each historical response voice, each historical recognition operation and each historical response service in the voice log data may be quantitatively counted to obtain an interaction number statistical result.

[0037] Specifically, a quantitative statistics is performed on at least one of each historical response voice, each historical recognition operation and each historical response service, which may include at least one of the following: determining various historical response voices under all historical response voices, counting the number of voices or the number of triggered devices for each historical response voice, and obtaining the statistical results of the number of interactions; determining various historical recognition operations under all historical recognition operations, counting the number of voices or the number of triggered devices for each historical recognition operation, and obtaining the statistical results of the number of interactions; determining various historical response services under all historical response services, counting the number of voices or the number of triggered devices for each historical response service, and obtaining the statistical results of the number of interactions.

[0038] Among them, the interaction number statistics result can be the number statistics result for at least one of each historical response voice, each historical recognition operation and each historical response service, or it can also be the number statistics result for at least one of each historical response voice, each historical recognition operation and each historical response service.

[0039] In a specific implementation, at least one of each historical response voice, each historical recognition operation, and each historical response service in the voice log data is counted to obtain an interaction number statistical result, including at least one of the following:

[0040] For each historical response voice, based on the number of voices or the number of triggering devices of the historical response voice, determine the number of voice interactions corresponding to the historical response voice;

[0041] For each historical recognition operation, based on the number of voices or the number of triggering devices of each historical response voice corresponding to the historical recognition operation, determine the number of operation interactions corresponding to the historical recognition operation;

[0042] For each historical response service, the number of service interactions corresponding to the historical response service is determined based on the number of voices or the number of triggering devices of each historical response voice corresponding to the historical response service.

[0043] In the above implementation, a quantitative statistical analysis can be performed on all historical response voices in the voice log data to obtain the number of voice interactions corresponding to each historical response voice. The number of voice interactions of each historical response voice can be the number of voices of the historical response voices of this type, or the number of voice interactions of each historical response voice can also be the number of triggering devices of the historical response voices of this type, that is, the number of vehicle computers that receive the historical response voices of this type.

[0044] For example, the number of voices corresponding to “I’m hot” is 1111, and the number of corresponding trigger devices is 253; the number of voices corresponding to “turn on the air conditioner” is 123, and the number of corresponding trigger devices is 17.

[0045] Alternatively, a quantitative statistical analysis may be performed on all historical recognition operations in the voice log data to obtain the number of operation interactions corresponding to each historical recognition operation. The number of operation interactions for each historical recognition operation may be the number of historical response voices corresponding to the historical recognition operation of that type, or the number of operation interactions for each historical recognition operation may be the number of triggering devices for the historical response voices corresponding to the historical recognition operation of that type.

[0046] For example, the number of voices corresponding to "navigation" is 5478, and the number of corresponding trigger devices is 369; the number of voices corresponding to "body control" is 3766, and the number of corresponding trigger devices is 89.

[0047] Alternatively, a quantitative statistical analysis may be performed on all historical response services in the voice log data to obtain the number of service interactions corresponding to each historical response service. The number of service interactions for each historical response service may be the number of voices of the historical response voices corresponding to the historical response services of the type, or the number of service interactions for each historical response service may be the number of triggering devices of the historical response voices corresponding to the historical response services of the type.

[0048] For example, the number of voices corresponding to "map" is 12661611, and the corresponding trigger device processing is 787; the number of voices corresponding to "vehicle control" is 26247, and the corresponding trigger device processing is 97.

[0049] In the above implementation, the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, or the number of service interactions corresponding to various historical response services can be determined through a quantity statistics statement (such as "topic:recorder.stream.start").

[0050] By determining at least one of the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services, statistical analysis of voice log data is achieved. Furthermore, the statistical results can provide data support for subsequent service development or optimization analysis.

[0051] It should be noted that this embodiment can perform quantitative statistics on all voice log data, and can also perform quantitative statistics on voice log data of different models separately to obtain statistical results of the number of interactions corresponding to different models; or, can also perform quantitative statistics on voice log data of different user attributes separately to obtain statistical results of the number of interactions corresponding to different user attributes.

[0052] Optionally, a quantitative analysis is performed on each historical response voice, each historical recognition operation, and at least one of each historical response service in the voice log data to obtain an interaction number statistical result, including: determining the trigger device type corresponding to each historical response voice; for each trigger device type, a quantitative analysis is performed on each historical response voice, each historical recognition operation, and at least one of each historical response service under the trigger device type to obtain an interaction number statistical result corresponding to the trigger device type.

[0053] Specifically, the number of voice interactions of various historical response voices under each trigger device type, or the number of operation interactions of various historical recognition operations, or the number of service interactions of various historical response services can be counted.

[0054] The benefit of separately counting the voice log data of different models or different user attributes is that the voice response services used by users of different models and different attributes may be quite different, such as the voice response services used by trucks (more used for navigation) and cars (more used for music), the voice response services used by off-road vehicles (more used for music) and electric scooters (more used for vehicle settings and vehicle control), and the voice response services used by young users (more used for jokes) and elderly users (more used for navigation). Therefore, in order to develop and optimize targeted vehicle voice services for different models and different users, the voice log data of different models or different user attributes can also be counted separately to refine the statistical results of the number of interactions for each model and each user, and further provide data support for subsequent targeted service development or optimization analysis.

[0055] Optionally, the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services can be sent to a business system for analysis, such as a reporting system, a machine learning system, a service recommendation system, or a user portrait system.

[0056] The method for processing vehicle-machine voice interaction data provided in the present embodiment obtains voice log data including each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice, and then performs quantitative statistics on each historical response voice, each historical recognition operation, and at least one of each historical response service, to obtain at least one of the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services, thereby realizing statistics on each vehicle-machine voice of successful interaction. The statistical results can be used for function development or function optimization without manual research, thus solving the problem that manual research consumes a lot of costs. In addition, the statistical results of successful interactions determined based on historical voice interaction data can reflect the actual usage of users. The development direction or optimization direction determined by the statistical results is highly accurate, thus avoiding the situation where the optimization development direction is wrong.

[0057] Figure 2 A flowchart of another method for processing vehicle-machine voice interaction data provided by an embodiment of the present application. Based on the above embodiments, optionally, an exemplary description is given of the process of determining the interaction results to be analyzed and / or the interaction results to be optimized based on the interaction number statistics. Figure 2 As shown, the method may specifically include the following steps:

[0058] S210. Acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice.

[0059] S220, performing quantitative statistics on at least one of each historical response voice, each historical recognition operation, and each historical response service in the voice log data to obtain an interaction number statistical result.

[0060] S230: Sort the interaction number statistics results to obtain a sorting result, wherein the sorting result includes: a first sorting result from large to small in number or a second sorting result from small to large in number.

[0061] Specifically, after determining the interaction number statistical results, the interaction number statistical results may be sorted to further determine the interaction results to be analyzed and / or the interaction results to be optimized.

[0062] In this embodiment, the various historical response voices, various historical recognition operations or various historical response services in the interaction number statistics results can be sorted in order from large to small to obtain a first sorting result; or, the various historical response voices, various historical recognition operations or various historical response services in the interaction number statistics results can be sorted in order from small to large to obtain a second sorting result.

[0063] S240, select at least one of the historical response voices, historical recognition operations, and historical response services before the first preset threshold in the first sorting result, or select at least one of the historical response voices, historical recognition operations, and historical response services after the first preset threshold in the second sorting result, as the interaction result to be analyzed; and / or, select at least one of the historical response voices, historical recognition operations, and historical response services before the second preset threshold in the second sorting result, or select at least one of the historical response voices, historical recognition operations, and historical response services after the second preset threshold in the first sorting result, as the interaction result to be optimized.

[0064] Among them, the interaction results to be analyzed are voices, operations or services that can be further developed, including at least one of popular response voices, popular recognition operations and popular response services; the interaction results to be optimized are voices, operations or services that can be further optimized, including at least one of unpopular response voices, unpopular recognition operations and unpopular response services.

[0065] Specifically, the first N1 (preset first threshold) historical response voices, historical recognition operations or historical response services can be selected from the first sorting result as the interaction results to be analyzed; or the last N1 (preset first threshold) historical response voices, historical recognition operations or historical response services can be selected from the second sorting result as the interaction results to be analyzed. In this way, the determination of popular response voices, popular recognition operations or popular response services can be achieved.

[0066] In addition, the first N2 (preset second threshold) historical response voices, historical recognition operations or historical response services can be selected from the second sorting results as the interaction results to be optimized; or, the last N2 (preset second threshold) historical response voices, historical recognition operations or historical response services can be selected from the first sorting results as the interaction results to be optimized.

[0067] It should be noted that in addition to directly determining the interaction results to be analyzed and / or the interaction results to be optimized based on the statistical results of the number of interactions, the present embodiment can also determine the interaction results to be optimized and / or the interaction results to be analyzed corresponding to each trigger device type, and then perform differentiated optimization on each trigger device type to improve the optimization accuracy of each vehicle model.

[0068] S250: Determine corresponding processed data output according to the interaction results to be analyzed and / or the interaction results to be optimized.

[0069] Specifically, the interaction results to be analyzed and / or the interaction results to be optimized may be directly output as processed data and sent to a preset interface of the cloud for display, or sent to other business systems.

[0070] In an optional embodiment, the corresponding processed data output is determined according to the interaction results to be analyzed and / or the interaction results to be optimized, including: if the interaction results to be analyzed are determined, and the interaction results to be analyzed include various historical response voices, then the historical response voices in the interaction results to be analyzed are determined as target analysis voices; based on each target analysis voice, a target knowledge graph is constructed, and the target knowledge graph is output; wherein, the target knowledge graph includes each target analysis voice and target display information related to each target analysis voice, and the target knowledge graph is used to determine the current display information corresponding to the current voice when the user's current voice is detected.

[0071] That is, if the determined interaction results to be analyzed include popular response voices, the historical response voices in the interaction results to be analyzed can be used as target analysis voices. Furthermore, for each target analysis voice, target display information related to the target analysis voice is obtained, and a target knowledge graph is constructed based on each target analysis voice and the target display information related to each target analysis voice. Among them, the target display information can be display information associated with the target analysis voice that helps users understand the vehicle functions; for example, information displayed in the form of pictures, text or videos.

[0072] After constructing the target knowledge graph, the target knowledge graph can be sent to each car computer. When the car computer detects the user's current voice, it can determine the current display information related to the current voice through the target knowledge graph and display the current display information on the car computer interface.

[0073] Exemplarily, the target knowledge graph may include popular speech 1 (lowering the cabin height) and target display information related to the popular speech 1 (graphic display of the location of the cabin height adjustment button and the adjustment steps); the target knowledge graph may also include popular speech 2 (opening the sunroof) and target display information related to the popular speech 2 (graphic display of the sunroof opening button and the opening size adjustment process).

[0074] In the above implementation, by analyzing the historical response voices with a large number of interactions, a target knowledge graph is constructed for the historical response voices with a large number of interactions, that is, the popular voices, so that when the user initiates the popular voice, the target knowledge graph can actively provide the user with related display information, further helping the user to understand the various vehicle functions, providing convenience for the user's vehicle use, and improving the user experience.

[0075] In another optional embodiment, the corresponding processed data output is determined according to the interaction results to be analyzed and / or the interaction results to be optimized, including: if the interaction results to be analyzed are determined, and the interaction results to be analyzed include various historical recognition operations, then the historical recognition operations in the interaction results to be analyzed are determined as target analysis operations; user historical input data corresponding to each target analysis operation is obtained, wherein the user historical input data includes at least one of vehicle-machine interface input data, vehicle control data, and user voice input data; user portraits corresponding to each preset user type are constructed based on the user historical input data, so as to determine the current type corresponding to the user through each user portrait and the user's current input data, and determine the preset recommended service corresponding to the current type; and output the preset recommended service to the user's terminal.

[0076] That is, if the determined interaction results to be analyzed include popular recognition operations, the historical recognition operations in the interaction results to be analyzed can be used as target analysis operations. Furthermore, for each target analysis operation, historical user input data such as vehicle interface input data, vehicle control data, or user voice input data are obtained. Among them, the vehicle interface input data may include information entered by the user on the vehicle interface and control information triggered by the user; the vehicle control data may include the user's driving control data or vehicle control data of the vehicle.

[0077] Specifically, based on the user's historical input data, user portraits corresponding to each preset user type can be constructed. Among them, the preset user type can be, for example, music type, entertainment type, navigation type, or car control type. The user portrait may include various tags and the weights corresponding to each tag. In this embodiment, the user portrait may include various user historical input data and the weights corresponding to various user historical input data. Exemplarily, in the user portrait corresponding to the music type, the weight corresponding to the operation of clicking to play music is 0.2, the weight corresponding to the voice initiating the music playback is 0.3, the weight corresponding to the voice initiating the search for singers is 0.45, the weight corresponding to opening the weather is 0.01, and so on.

[0078] After constructing user portraits corresponding to each preset user type, they can be sent to the vehicle computer, which can then calculate the reference weight of each user portrait through the user's current input data and the weight corresponding to each current input data in the user portrait, and then determine the preset user type corresponding to the user portrait with the largest reference weight as the current type.

[0079] Furthermore, after determining the current type of the user, the corresponding preset recommended service can be determined according to the current user type, and the preset recommended service can be output to the user's terminal (such as a car computer, mobile phone, etc.) to actively recommend the preset recommended service corresponding to the current type to the user's terminal.

[0080] Among them, the preset recommendation service can be a pre-set push service related to the preset user type. For example, the preset recommendation service corresponding to the music type can be a music service, the preset recommendation service corresponding to the entertainment type can be a joke service or a poetry service, and the preset recommendation service corresponding to the vehicle control type can be a vehicle setting and control service.

[0081] Through the above method, user portraits of various preset user types can be constructed according to popular recognition operations, and then the current type of the user can be determined according to the user's current input data and each user portrait, and the preset recommended service corresponding to the current type can be output to the user's terminal, thereby realizing user portrait construction and active service recommendation, helping users to understand the in-vehicle voice service, and then facilitating users to use the in-vehicle voice service, further improving the user experience.

[0082] Of course, in addition to constructing user portraits based on the user historical input data corresponding to the target analysis operation as mentioned above, you can also obtain the user historical input data corresponding to the target analysis voice or target analysis service, and construct user portraits corresponding to each preset user type based on the user historical input data corresponding to the target analysis voice or target analysis service.

[0083] In another optional implementation, the corresponding processed data output is determined based on the interaction results to be analyzed and / or the interaction results to be optimized, including: if the interaction results to be analyzed are determined, and the interaction results to be analyzed include various historical response services, then the historical response services in the interaction results to be analyzed are determined as target analysis services; when it is detected that the user has not triggered each target analysis service, each target analysis service is output to the user's terminal.

[0084] That is, if the determined interaction results to be analyzed include popular response services, the historical response services in the interaction results to be analyzed can be used as target analysis services. Further, if it is detected that the user has not triggered the target analysis service, each target analysis service can be output to the user's terminal to actively recommend each target analysis service to the user.

[0085] Exemplarily, outputting each target analysis service to the user's terminal may be: playing a guidance service voice corresponding to the target analysis service, wherein the guidance service voice is used to indicate the use of the target analysis service.

[0086] For example, when the target analysis service is music, if it is detected that the user has not triggered the service, "You can say to me: play a song" or "You can say to me: play a piece of music to relax" can be actively played. Of course, outputting the preset recommendation service to the user's terminal can also be achieved by playing the guidance service voice corresponding to the preset recommendation service.

[0087] Through the above method, popular response services can be determined, and when the user has not triggered a popular response service, the popular response service can be actively recommended to the user to help the user understand the in-vehicle voice service, thereby facilitating the user to use the in-vehicle voice service and further improving the user experience.

[0088] In another optional implementation, the corresponding processed data output is determined based on the interaction results to be analyzed and / or the interaction results to be optimized, including: if the interaction results to be optimized are determined, and the interaction results to be optimized include various historical response services, then the historical response services in the interaction results to be optimized are determined as target optimization services; online upgrade files corresponding to each target optimization service are determined, and the online upgrade files are output to the vehicle computers corresponding to each target optimization service, so that each vehicle computer can eliminate the target optimization service through the online upgrade files.

[0089] That is, if the determined interaction results to be optimized include unpopular response services, the historical response services in the interaction results to be optimized can be used as target optimization services. Furthermore, the online upgrade files corresponding to each target optimization service are determined, wherein the online upgrade files are used to eliminate each target optimization service in the vehicle computer. Specifically, the online upgrade files can be actively sent to the vehicle computer corresponding to each target optimization service, or the vehicle computer corresponding to each target optimization service can actively obtain the online upgrade files, and then the vehicle computer eliminates each target optimization service by running the online upgrade files. Among them, the vehicle computer corresponding to each target optimization service can be a vehicle computer equipped with each target optimization service.

[0090] Through the above method, the response services with lower usage frequency can be removed from the car computer. Without affecting the provision of voice services to users, the load required for the car computer voice service can be reduced, the efficiency of the car computer voice service response can be improved, and cost reduction and efficiency improvement can be achieved.

[0091] It should be noted that in the above-mentioned implementation methods of constructing the target knowledge graph, constructing the user portrait, determining the target analysis service and determining the target optimization service, it is also possible to determine the target knowledge graph, user portrait, target analysis service or target optimization service under each trigger device type according to the interaction results to be analyzed and / or the interaction results to be optimized corresponding to each trigger device type, thereby achieving targeted optimization of each trigger device type and further improving the optimization accuracy.

[0092] The method for processing vehicle-machine voice interaction data provided in this embodiment obtains at least one of the interaction results to be analyzed and the interaction results to be optimized by sorting the statistical results of the number of interactions, thereby realizing the determination of at least one of the popular response voices, popular recognition operations, popular response services, unpopular response voices, unpopular recognition operations and unpopular response services, providing data support for the mining of new scenarios of vehicle-machine voice, facilitating subsequent targeted optimization, and also achieving cost reduction and efficiency improvement.

[0093] For example, Figure 3 As shown, a processing process of vehicle-machine voice interaction data is shown. Among them, the voice server can collect and store the voice log data of the test environment and the formal environment respectively, and send the voice log data to Kafka. Further, Kafka sends the voice log data to HIVE / Hadoop / scheduler, and HIVE / Hadoop / scheduler counts the voice log data through RDS / MySQL according to the user's interaction statistics instructions, obtains the interaction number statistics results, sorts the interaction number statistics results, and determines the interaction results to be analyzed and / or the interaction results to be optimized based on the sorting results.

[0094] Furthermore, the interaction results to be analyzed and / or the interaction results to be optimized can be displayed. Alternatively, the interaction results to be analyzed and / or the interaction results to be optimized are sent to a reporting system to conduct user behavior research through the reporting system and explore new scenarios for vehicle-based voice services. Alternatively, further machine learning is performed to construct a target knowledge graph through machine learning and explore the vehicle-understanding master function of the vehicle-based system. Alternatively, it is sent to a recommendation system to recommend popular response services to users through the recommendation system, such as setting voice guide words corresponding to popular response services ("You can say this to me: XX"). Alternatively, user portraits are further constructed to construct user portraits corresponding to each user type.

[0095] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0096] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a device for processing vehicle-machine voice interaction data. Figure 4 The structure and operation of a device for processing vehicle-machine voice interaction data provided in an embodiment of the present application are shown.

[0098] refer to Figure 4 , the vehicle-machine voice interaction data processing device includes a log acquisition module 410 and a log statistics module 420, wherein;

[0099] The log acquisition module 410 is used to acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice;

[0100] The log statistics module 420 is used to perform quantitative statistics on at least one of the historical response voices, the historical recognition operations and the historical response services in the voice log data to obtain an interaction number statistical result;

[0101] The interaction number statistics include at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services.

[0102] The processing device for vehicle-machine voice interaction data provided in the embodiment of the present application obtains voice log data including each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice, and then performs quantitative statistics on each historical response voice, each historical recognition operation, and at least one of each historical response service, to obtain at least one of the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services, thereby realizing statistics on each vehicle-machine voice of successful interaction. The statistical results can be used for function development or function optimization without manual research, thus solving the problem that manual research requires a lot of cost. In addition, the statistical results of successful interactions determined based on historical voice interaction data can reflect the actual usage of users. The development direction or optimization direction determined by the statistical results is highly accurate, thus avoiding the situation where the optimization development direction is wrong.

[0103] On the basis of the above implementation mode, optionally, the processing device of the vehicle-machine voice interaction data also includes a sorting module, which is used to sort the statistical results of the number of interactions to obtain a sorting result, wherein the sorting result includes: a first sorting result in order of number from large to small or a second sorting result in order of number from small to large; selecting at least one of the historical response voices, historical recognition operations and historical response services before a first threshold value preset in the first sorting result, or selecting at least one of the historical response voices, historical recognition operations and historical response services after a first threshold value preset in the second sorting result, as the interaction result to be analyzed; and / or selecting at least one of the historical response voices, historical recognition operations and historical response services before a second threshold value preset in the second sorting result, or selecting at least one of the historical response voices, historical recognition operations and historical response services after a second threshold value preset in the first sorting result, as the interaction result to be optimized; determining the corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized.

[0104] On the basis of the above implementation mode, optionally, the sorting module also includes a knowledge graph construction unit, and the knowledge graph construction unit is used to, if the interaction result to be analyzed is determined, and the interaction result to be analyzed includes various historical response voices, then determine the historical response voice in the interaction result to be analyzed as the target analysis voice; construct a target knowledge graph based on each of the target analysis voices, and output the target knowledge graph; wherein, the target knowledge graph includes each of the target analysis voices and target display information related to each of the target analysis voices, and the target knowledge graph is used to determine the current display information corresponding to the current voice when the user's current voice is detected.

[0105] On the basis of the above implementation mode, optionally, the sorting module also includes a user portrait construction unit, and the user portrait construction unit is used to, if the interaction result to be analyzed is determined, and the interaction result to be analyzed includes various historical identification operations, then determine the historical identification operations in the interaction result to be analyzed as target analysis operations; obtain user historical input data corresponding to each target analysis operation, wherein the user historical input data includes at least one of vehicle-machine interface input data, vehicle control data, and user voice input data; construct user portraits corresponding to each preset user type based on the user historical input data, determine the current type corresponding to the user through each of the user portraits and the user's current input data, and determine the preset recommended service corresponding to the current type; and output the preset recommended service to the user's terminal.

[0106] On the basis of the above implementation mode, optionally, the sorting module also includes a service recommendation unit, and the service recommendation unit is used to, if the interaction result to be analyzed is determined and the interaction result to be analyzed includes various historical response services, determine the historical response service in the interaction result to be analyzed as the target analysis service; when it is detected that the user has not triggered each of the target analysis services, output each of the target analysis services to the user's terminal.

[0107] On the basis of the above implementation, optionally, the sorting module also includes a service elimination unit, and the service elimination unit is used to, if the interaction result to be optimized is determined, and the interaction result to be optimized includes various historical response services, then determine the historical response service in the interaction result to be optimized as the target optimization service; determine the online upgrade file corresponding to each of the target optimization services, and output the online upgrade file to the vehicle computer corresponding to each of the target optimization services, so that each of the vehicle computers can eliminate the target optimization service through the online upgrade file.

[0108] On the basis of the above implementation mode, optionally, the log statistics module 420 is also used to perform at least one of the following: for each of the historical response voices, based on the number of voices of the historical response voices or the number of triggering devices, determining the number of voice interactions corresponding to the historical response voice; for each of the historical recognition operations, based on the number of voices of each historical response voice or the number of triggering devices corresponding to the historical recognition operation, determining the number of operation interactions corresponding to the historical recognition operation; for each of the historical response services, based on the number of voices of each historical response voice or the number of triggering devices corresponding to the historical response service, determining the number of service interactions corresponding to the historical response service.

[0109] On the basis of the above implementation mode, optionally, the log statistics module 420 is also used to determine the trigger device type corresponding to each of the historical response voices; for each trigger device type, the historical response voices, the historical recognition operations and at least one of the historical response services under the trigger device type are quantitatively counted to obtain the interaction number statistics corresponding to the trigger device type.

[0110] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0111] The device of the above embodiment is used to implement the corresponding vehicle-machine voice interaction data processing method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0112] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments, the present application also provides an electronic device, including 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 method for processing vehicle-machine voice interaction data described in any of the above embodiments is implemented.

[0113] Figure 5 The structure and operation of a more specific electronic device provided by this embodiment are shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040 and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030 and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0114] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0115] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0116] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0117] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0118] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0119] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0120] The electronic device of the above-mentioned embodiment is used to implement the corresponding vehicle-machine voice interaction data processing method in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0121] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method for processing vehicle-machine voice interaction data as described in any of the above embodiments.

[0122] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0123] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the method for processing vehicle-machine voice interaction data as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0124] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0125] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0126] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0127] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A method for processing vehicle-machine voice interaction data, characterized in that: include: Acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice; Performing quantitative statistics on at least one of each of the historical response voices, each of the historical recognition operations, and each of the historical response services in the voice log data to obtain an interaction number statistical result, wherein the interaction number statistical result includes at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services; Sorting the interaction number statistical results to obtain a sorting result, wherein the sorting result includes: a first sorting result in which the number is from large to small or a second sorting result in which the number is from small to large; selecting at least one of the historical response voices, historical recognition operations, and historical response services before a first threshold value preset in the first sorting result, or selecting at least one of the historical response voices, historical recognition operations, and historical response services after a first threshold value preset in the second sorting result as the interaction result to be analyzed; and / or, Selecting at least one of the historical response voices, historical recognition operations, and historical response services before a second threshold value preset in the second sorting result, or selecting at least one of the historical response voices, historical recognition operations, and historical response services after a second threshold value preset in the first sorting result as the interaction result to be optimized; Determine the corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized; Wherein, determining the corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized includes: If the interaction result to be analyzed is determined, and the interaction result to be analyzed includes each historical response voice, then the historical response voice in the interaction result to be analyzed is determined as the target analysis voice; Construct a target knowledge graph based on the target analysis speech, and output the target knowledge graph; The target knowledge graph includes each of the target analysis voices and target display information related to each of the target analysis voices, and the target knowledge graph is used to determine the current display information corresponding to the current voice when the user's current voice is detected; Wherein, determining the corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized includes: If the interaction result to be optimized is determined, and the interaction result to be optimized includes various historical response services, determining the historical response service in the interaction result to be optimized as the target optimization service; An online upgrade file corresponding to each of the target optimization services is determined, and the online upgrade file is output to a vehicle computer corresponding to each of the target optimization services, so that each of the vehicle computers can remove the target optimization service through the online upgrade file.

2. The method according to claim 1, characterized in that The determining corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized includes: If the interaction result to be analyzed is determined, and the interaction result to be analyzed includes various historical identification operations, determining the historical identification operations in the interaction result to be analyzed as target analysis operations; Acquire user historical input data corresponding to each of the target analysis operations, wherein the user historical input data includes at least one of vehicle interface input data, vehicle control data, and user voice input data; Constructing user portraits corresponding to respective preset user types based on the user's historical input data, determining the current type corresponding to the user through the user portraits and the user's current input data, and determining the preset recommended service corresponding to the current type; The preset recommended service is output to the user's terminal.

3. The method according to claim 1, characterized in that The determining corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized includes: If the interaction result to be analyzed is determined, and the interaction result to be analyzed includes various historical response services, determining the historical response service in the interaction result to be analyzed as a target analysis service; When it is detected that the user has not triggered each of the target analysis services, each of the target analysis services is output to the terminal of the user.

4. The method according to any one of claims 1 to 3, characterized in that The performing of quantitative statistics on at least one of each of the historical response voices, each of the historical recognition operations, and each of the historical response services in the voice log data to obtain an interaction number statistical result includes at least one of the following: For each of the historical response voices, based on the number of voices or the number of triggering devices of the historical response voices, determining the number of voice interactions corresponding to the historical response voices; For each of the historical recognition operations, determining the number of operation interactions corresponding to the historical recognition operation based on the number of voices or the number of triggering devices of each historical response voice corresponding to the historical recognition operation; For each of the historical response services, the number of service interactions corresponding to the historical response service is determined based on the number of voices or the number of triggering devices of each historical response voice corresponding to the historical response service.

5. The method according to claim 1, characterized in that The method of performing quantitative statistics on at least one of the historical response voices, the historical recognition operations, and the historical response services in the voice log data to obtain the interaction number statistical result includes: Determine the trigger device type corresponding to each of the historical response voices; For each trigger device type, at least one of each historical response voice, each historical recognition operation and each historical response service under the trigger device type is counted to obtain a statistical result of the number of interactions corresponding to the trigger device type.

6. A device for processing vehicle-machine voice interaction data, characterized in that: include: A log acquisition module, used to acquire voice log data, wherein the voice log data includes each historical response voice of successful interaction, each historical recognition operation corresponding to each historical response voice, and each historical response service corresponding to each historical response voice; A log statistics module, used to perform quantitative statistics on at least one of the historical response voices, the historical recognition operations, and the historical response services in the voice log data to obtain an interaction number statistical result; The interaction number statistics result includes at least one of the following: the number of voice interactions corresponding to various historical response voices, the number of operation interactions corresponding to various historical recognition operations, and the number of service interactions corresponding to various historical response services; The processing device for vehicle-machine voice interaction data also includes a sorting module, which is used to sort the statistical results of the number of interactions to obtain a sorting result, wherein the sorting result includes: a first sorting result with a large number or a second sorting result with a small number; selecting at least one of the historical response voices, historical recognition operations and historical response services before a first threshold value preset in the first sorting result, or selecting at least one of the historical response voices, historical recognition operations and historical response services after a first threshold value preset in the second sorting result, as the interaction result to be analyzed; and / or selecting at least one of the historical response voices, historical recognition operations and historical response services before a second threshold value preset in the second sorting result, or selecting at least one of the historical response voices, historical recognition operations and historical response services after a second threshold value preset in the first sorting result, as the interaction result to be optimized; determining the corresponding processed data output according to the interaction result to be analyzed and / or the interaction result to be optimized; The sorting module also includes a knowledge graph construction unit, and the knowledge graph construction unit is used to determine the historical response voice in the interaction result to be analyzed as the target analysis voice if the interaction result to be analyzed is determined and the interaction result to be analyzed includes various historical response voices; construct a target knowledge graph based on each of the target analysis voices, and output the target knowledge graph; wherein the target knowledge graph includes each of the target analysis voices and target display information related to each of the target analysis voices, and the target knowledge graph is used to determine the current display information corresponding to the current voice when the user's current voice is detected; The sorting module also includes a service recommendation unit, which is used to, if the interaction result to be analyzed is determined and the interaction result to be analyzed includes various historical response services, determine the historical response service in the interaction result to be analyzed as the target analysis service; when it is detected that the user has not triggered each of the target analysis services, output each of the target analysis services to the user's terminal.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for processing vehicle-machine voice interaction data as described in any one of claims 1 to 5 is implemented.

Images