Task information identification method and device, equipment and storage medium

Abstract

The invention discloses a task information identification method and device, equipment and a storage medium, and relates to the field of vehicle-mounted technology, and the method comprises the steps: obtaining voice information, inputting the voice information into a preset deep learning model, obtaining an intention identification result, a vehicle-mounted use intention and slot position information, combining the intention identification result according to the vehicle-mounted use intention and the slot position information, and obtaining a task information identification result; obtaining an extended intention recognition result, and obtaining task information according to the extended intention recognition result. The voice information is integrated through the deep learning model, intention and slot position recognition is realized, an accurate task instruction is generated, the accuracy of voice recognition is improved, and the user experience is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-mounted technology, and in particular to a task information recognition method, device, equipment and storage medium. Background Art

[0002] With the development of smart car technology, in-vehicle systems are increasingly integrating voice recognition functions to provide a safer and more convenient way of interaction. Users can control navigation, media playback, communication and other in-vehicle services through voice commands. This interactive method reduces the driver's reliance on physical controls during driving, thereby helping to improve driving safety. However, in order to achieve efficient and accurate voice recognition, the system needs to be able to understand and process complex language structures and diverse user intentions.

[0003] The current in-vehicle task recognition collects the user's voice commands through the in-vehicle microphone, performs noise reduction and enhancement processing on the collected voice signals, converts the pre-processed voice signals into text information, analyzes the recognized text to determine the user's intentions and related parameters, and finally executes the corresponding control commands based on the parsed intentions and slot information.

[0004] Although existing technologies have made some progress, they still have obvious shortcomings. First, speech recognition and text processing tasks are usually completed by independent modules, resulting in a large gap between modules and increasing communication and coordination costs. Moreover, in a noisy in-car environment or when faced with voice input with a local accent, it is difficult to accurately recognize and understand the user's intention. Therefore, there is an urgent need for a more efficient and accurate method to improve the quality of speech recognition. Summary of the invention

[0005] The main purpose of this application is to provide a task information recognition method, device, equipment and storage medium, aiming to solve the technical problem of how to improve the accuracy of speech recognition task extraction.

[0006] To achieve the above objectives, the present application proposes a task information identification method, the method comprising:

[0007] Get voice information;

[0008] Inputting the voice information into a preset deep learning model to obtain an intention recognition result, vehicle-mounted usage intention, and slot information, wherein the preset deep learning model at least includes a voice encoder, a predictor, and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer;

[0009] Obtaining an expanded intention recognition result according to the vehicle-mounted use intention and the slot information combined with the intention recognition result;

[0010] The task information is obtained according to the expanded intention recognition result.

[0011] In one embodiment, the steps of inputting a preset deep learning model according to the voice information to obtain the intention recognition result, the vehicle use intention and the slot information include:

[0012] Inputting the speech information into a preset deep learning model, converting the speech information through the speech encoder to obtain a feature vector;

[0013] Inputting the feature vector into the predictor to obtain predicted text information;

[0014] Inputting the predicted text information into the joint network layer for integration to generate an intent recognition result;

[0015] An analysis is performed based on the intention recognition result to obtain vehicle usage intention and slot information.

[0016] In one embodiment, the step of inputting the predicted text information into the joint network layer for integration to generate an intent recognition result includes:

[0017] Inputting the predicted text information into the joint network layer;

[0018] The predicted text information is integrated through the joint network layer to perform nonlinear changes and fusion operations to obtain the intent recognition result.

[0019] In one embodiment, the step of obtaining an expanded intention recognition result based on the vehicle-mounted use intention and the slot information in combination with the intention recognition result includes:

[0020] Expanding the vehicle usage intention and the slot information to obtain an extended text vocabulary, wherein the vehicle usage intention at least includes navigation, media control, air conditioning adjustment, and seat control, and the slot information at least includes a geographic location, a song name, and a contact;

[0021] The extended text vocabulary is expanded to the intention recognition result to obtain an expanded intention recognition result.

[0022] In one embodiment, before the step of expanding the extended text vocabulary to the intention recognition result to obtain the extended intention recognition result, the step includes:

[0023] Obtaining intent slot information according to the extended text vocabulary;

[0024] Setting the intent slot information to a preset label;

[0025] The step of expanding the extended text vocabulary to the intention recognition result to obtain an extended intention recognition result includes:

[0026] When the preset tag is detected in the intention recognition result, the intention recognition result is expanded in combination with the extended text vocabulary to obtain an expanded intention recognition result.

[0027] In one embodiment, the step of obtaining task information according to the expanded intention recognition result includes:

[0028] Analyze the expanded intent recognition result to obtain the user intent and the slot information corresponding to the user intent;

[0029] The task information is obtained by performing conversion according to the user intention and the slot information.

[0030] In one embodiment, after the step of converting according to the user intention and the relevant slot information to obtain the task information, the following steps are included:

[0031] A task instruction is generated according to the task information and sent to the vehicle-mounted device, so that the vehicle-mounted device receives the task instruction and performs corresponding operations.

[0032] In addition, to achieve the above purpose, the present application also proposes a task information identification device, the task information identification device comprising:

[0033] An acquisition module, used for acquiring voice information;

[0034] The acquisition module is further used to input the voice information into a preset deep learning model to obtain an intention recognition result, vehicle-mounted use intention and slot information, wherein the preset deep learning model at least includes a voice encoder, a predictor and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer;

[0035] An expansion module, used to obtain an expanded intention recognition result according to the vehicle-mounted use intention and the slot information in combination with the intention recognition result;

[0036] The analysis module is used to obtain task information according to the expanded intention recognition result.

[0037] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable medium, and a computer program is stored on the medium. When the computer program is executed by a processor, the steps of the task information identification method described above are implemented.

[0038] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the task information identification method described above are implemented.

[0039] This application obtains voice information, inputs the voice information into a preset deep learning model, obtains intent recognition results, vehicle-mounted use intentions and slot information, combines the intent recognition results with the vehicle-mounted use intentions and slot information, obtains expanded intent recognition results, and obtains task information based on the expanded intent recognition results. By integrating voice information through a deep learning model, intent and slot recognition are achieved, and precise task instructions are generated, which improves the accuracy of voice recognition and enhances user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0041] Figure 1 This is a flowchart of the first embodiment of the task information identification method of this application;

[0042] Figure 2 This is a flow chart of the second embodiment of the task information identification method of this application;

[0043] Figure 3 This is a schematic diagram of the module structure of the task information identification device according to an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the device structure of the hardware operating environment involved in the task information identification method in the embodiment of the present application;

[0045] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0047] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0048] The main solution of the embodiment of the present application is: by acquiring voice information, inputting the voice information into a preset deep learning model, obtaining intention recognition results, vehicle-mounted usage intentions and slot information, combining the intention recognition results with the vehicle-mounted usage intentions and slot information, obtaining expanded intention recognition results, and obtaining task information based on the expanded intention recognition results.

[0049] Based on this, the present application embodiment provides a task information identification method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the task information identification method of this application.

[0050] In this embodiment, the task information identification method includes steps S10 to S40:

[0051] Step S10, obtaining voice information.

[0052] It should be noted that the system captures the user's original voice commands through the vehicle microphone array, and the voice commands contain complex semantics and diverse intentions. In order to improve the accuracy of recognition, the collected original voice signal will undergo a series of preprocessing steps, including noise suppression, echo cancellation, volume normalization, etc., to ensure the quality of the voice data. Subsequently, the preprocessed voice data will be input into a preset deep learning model, which is a complex neural network, such as a convolutional neural network, a recurrent neural network, a transducer model or an attention mechanism model. In this embodiment, the deep learning model adopts a transducer model.

[0053] Step S20, input the voice information into a preset deep learning model to obtain intention recognition results, vehicle usage intention and slot information. The preset deep learning model includes at least a voice encoder, a predictor and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer.

[0054] It should be noted that the system inputs the collected voice information into the Transducer model, which contains three core components: speech encoder, predictor and joint network layer. The speech encoder is responsible for converting the speech signal into a high-dimensional feature representation that can be further processed. These features include not only the physical properties of the speech, but also the identity characteristics and emotional state of the speaker. The predictor uses high-dimensional features to predict possible text outputs. It gradually builds possible sentence sequences by learning the complex mapping relationship between speech and text. The most important is the joint network layer, which serves as a bridge between the speech encoder and the predictor, integrating the outputs of the two and fusing information. The joint network layer uses complex algorithms, such as attention mechanisms or memory networks, to analyze and understand the intent and slot information in the speech to obtain intent recognition results.

[0055] Furthermore, the voice information is input into the preset deep learning model, and the voice information is converted through the voice encoder to obtain the feature vector. The voice information is input into the Transducer model, in which the voice information is first sent to the voice encoder for conversion. As part of the Transducer model, the voice encoder is responsible for converting the input raw voice signal into multiple high-dimensional feature vectors. These feature vectors are not just simple digitization of the voice signal, but also contain rich information, such as basic attributes of speech such as pitch, timbre, rhythm, and can even capture the speaker's emotional characteristics, environmental background noise, and subtle changes in voice pronunciation.

[0056] Furthermore, step S20 specifically includes: inputting the feature vector into the predictor to obtain predicted text information. The feature vector is then input into the predictor, which, as a core component of the Transducer model, begins to generate predicted text information based on the received feature vector. By combining contextual information and symbols predicted based on the tone, an output symbol sequence is gradually constructed, and then integrated to obtain the predicted text information. The predictor uses recursive neural network technology, which means that it can effectively handle long-distance dependencies and maintain a high prediction accuracy even when faced with complex voice input. In addition, the predictor also has a certain degree of flexibility and can dynamically adjust its prediction strategy according to different voice input environments (such as speaking speed, accent differences, etc.).

[0057] Furthermore, step S20 specifically also includes: inputting the predicted text information into the joint network layer, integrating the predicted text information through the joint network layer to perform nonlinear changes and fusion operations, and obtain the intention recognition result. As an important component of the Transducer model, the joint network layer is responsible for comprehensive processing of the feature vectors from the speech encoder and the text information generated by the predictor. Specifically, the predicted text information is input into the joint network layer, and nonlinear changes and fusion operations are performed by integrating the predicted text information to obtain the intention recognition result. Through nonlinear transformation and multi-level fusion mechanisms, the joint network can make full use of speech features and contextual information to generate intention recognition results. This integration method not only ensures that the output of the system at each time step can accurately reflect the true text of the current speech signal, but also can extract the vehicle-mounted usage intention and slot information related to the speech segment.

[0058] Step S30, obtaining an expanded intention recognition result based on the vehicle usage intention and the slot information combined with the intention recognition result.

[0059] It is important to understand that vehicle usage intent refers to commands or requests issued by users through voice, which are intended to allow the vehicle to perform a certain function or service, such as navigating to a certain location, playing specific music, or adjusting the temperature in the car. Slot information refers to specific parameters or data required to complete these commands, such as the address of the destination, the name of the song, or the set temperature value. Based on the vehicle usage intent and slot information combined with the intent recognition result, the system will generate a more detailed expanded intent recognition result.

[0060] Furthermore, step S30 specifically includes: expanding the vehicle-mounted usage intention and slot information to obtain an extended text vocabulary. The vehicle-mounted usage intention and slot information are expanded into an extended text vocabulary so that the system can understand the user's voice commands more comprehensively. The extended text vocabulary not only includes basic words and phrases, but also specifically adds all the intention words related to vehicle-mounted functions, such as navigation, media control, air conditioning adjustment and seat control. At the same time, it also covers various specific slot information that users may mention when using these functions, such as the geographical location of the destination, the name of the song you want to play, or the name of the contact.

[0061] Furthermore, step S30 specifically also includes: expanding the extended text vocabulary to the intent recognition result to obtain an extended intent recognition result. This process involves combining the identified user intent with specific slot information. For example, if the user says "navigate to Square A", the system will not only recognize the action intention of "navigation", but also extract the geographical entity of "Square A" as slot information. These information together constitute a complete extended intent recognition result. In this way, the system not only understands what the user wants to do, but also clarifies the specific operation details that the user wants to perform.

[0062] Before obtaining the extended intent recognition result, the intent slot information is obtained according to the extended text vocabulary, and the intent slot information is set as a preset label. The intent slot information obtained according to the extended text vocabulary is set as a preset label. When the preset label is detected in the intent recognition result, the intent recognition result is expanded in combination with the extended text vocabulary to obtain the extended intent recognition result. When the system detects these preset labels during the intent recognition process, a process is automatically triggered to combine the information in the extended text vocabulary with the current intent recognition result to generate a more detailed extended intent recognition result. For example, if the user's voice command is "navigate to the nearest hospital", the system will recognize "navigation" as the intent and "nearest hospital" as the slot information. At this time, the system will check whether there is a preset label related to navigation. Once confirmed, the relevant vocabulary and information in the extended text vocabulary are used to supplement and improve the intent recognition result, so that the final generated extended intent recognition result not only contains the user's original intention, but also contains more specific operation details, such as the specific destination of navigation. This method, which comprehensively considers acoustic features, historical text, and historical slot information, ensures that when performing intent recognition, the system can more accurately grasp the user's true intention, thereby generating more accurate and consistent intent recognition results.

[0063] Step S40, obtaining task information according to the expanded intention recognition result.

[0064] It should be noted that the task information obtained based on the expanded intent recognition results refers to the system generating a specific executable operation guide by comprehensively analyzing the intent and slot information in the user's voice command. After obtaining the task information, it also includes generating task instructions based on the task information and sending them to the on-board device, so that the on-board device receives the task instructions and performs corresponding operations. For example, when the system recognizes that the user says "navigate to the nearest hospital", it not only understands the intention of "navigation", but also captures the geographical location slot of "nearest hospital". The system can further refine and convert it into actual task information, such as launching a navigation application, searching for nearby hospital locations, and loading the navigation path onto the on-board display for the driver's reference.

[0065] This embodiment provides a task information recognition method, which obtains voice information, inputs the voice information into a preset deep learning model, obtains intent recognition results, vehicle-mounted use intentions and slot information, combines the intent recognition results with the vehicle-mounted use intentions and slot information, obtains an expanded intent recognition result, and obtains task information based on the expanded intent recognition results. By integrating voice information through a deep learning model, intent and slot recognition are achieved, and precise task instructions are generated, thereby improving the accuracy of voice recognition and enhancing user experience.

[0066] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 The task information identification method step S40 further includes steps S401 to S402:

[0067] Step S401, analyzing the expanded intent recognition result to obtain the user intent and the slot information corresponding to the user intent.

[0068] It should be noted that this process involves in-depth analysis of the intent recognition results obtained from the deep learning model, especially the joint network layer, to extract the user's true intent and related slot information. The expanded intent recognition results usually contain richer semantic content and contextual information, which enables the system to understand the user's instructions more accurately. During the analysis process, the system uses natural language processing (NLP) techniques, such as semantic role labeling and dependency syntax analysis, to identify the components of the sentence and their relationships. For example, the system can recognize that "play" is an action intent, "jazz" is a music type slot, and "high volume" may represent an instruction to adjust the volume. In addition, the system may also use machine learning algorithms, such as classifiers or regression models, to predict and verify the accuracy of intents and slots.

[0069] Step S402: converting according to the user intention and the slot information to obtain task information.

[0070] It is important to understand that after obtaining the user's intent and related slot information, the system can convert it to generate more precise task instructions. During the conversion process, the system first verifies and calibrates the identified intent and slot information to ensure the accuracy of the information. For example, if the user says "I want to listen to the latest pop songs", the system not only recognizes the intent of "playing music", but also extracts the user's specific preferences from "latest" and "popular songs" and generates a playback task, thereby providing personalized music playback services.

[0071] Through this precise task information conversion, the vehicle system can provide users with more intelligent and personalized services. Users can control various functions of the vehicle, such as adjusting the air conditioning temperature, making calls, sending messages, etc., through simple voice commands, which greatly improves the convenience and safety of driving.

[0072] This embodiment analyzes the extended intent recognition results to obtain the user intent and related slot information, and converts the user intent and related slot information to obtain task information. By analyzing the extended intent recognition results, the user intent and slot information are accurately extracted and converted into specific task information, thereby improving accuracy.

[0073] This application also provides a task information identification device, please refer to Figure 3 , the device comprises:

[0074] The acquisition module 10 is used to acquire voice information.

[0075] The acquisition module 10 is also used to input the voice information into a preset deep learning model to obtain intention recognition results, vehicle usage intention and slot information. The preset deep learning model includes at least a voice encoder, a predictor and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer.

[0076] The expansion module 20 is used to obtain an expanded intention recognition result based on the vehicle-mounted use intention and the slot information in combination with the intention recognition result.

[0077] The analysis module 30 is used to obtain task information according to the expanded intention recognition result.

[0078] The task information identification device provided by the present application adopts the task information identification method in the above embodiment, which can solve the technical problem of how to improve the accuracy of speech recognition task extraction. Compared with the prior art, the beneficial effects of the task information identification device provided by the present application are the same as the beneficial effects of the task information identification method provided by the above embodiment, and the other technical features in the task information identification device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0079] In one embodiment, the acquisition module 10 is also used to input voice information into a preset deep learning model, convert the voice information through a voice encoder to obtain a feature vector; input the feature vector into a predictor to obtain predicted text information; integrate the predicted text information into a joint network layer to generate an intent recognition result; analyze the intent recognition result to obtain the vehicle-mounted usage intention and slot information.

[0080] In one embodiment, the acquisition module 10 is further used to input the predicted text information into the joint network layer; the predicted text information is integrated by the joint network layer to perform nonlinear changes and fusion operations to obtain the intent recognition result.

[0081] In one embodiment, the expansion module 20 is also used to expand the vehicle-mounted usage intention and slot information to obtain an extended text vocabulary, wherein the vehicle-mounted usage intention includes at least navigation, media control, air conditioning adjustment and seat control, and the slot information includes at least geographic location, song name and contact; the extended text vocabulary is expanded to the intention recognition result to obtain an extended intention recognition result.

[0082] In one embodiment, the expansion module 20 is further used to obtain the intention slot information according to the extended text vocabulary; and set the intention slot information as a preset label;

[0083] In one embodiment, the expansion module 20 is further used to expand the intent recognition result in combination with the extended text vocabulary to obtain an expanded intent recognition result when a preset tag is detected in the intent recognition result.

[0084] In one embodiment, the analysis module 30 is further used to analyze the expanded intention recognition result to obtain the user intention and the slot information corresponding to the user intention; and convert the user intention and the slot information to obtain the task information.

[0085] In one embodiment, the analysis module 30 is further used to generate a task instruction according to the task information, and send it to the vehicle-mounted device, so that the vehicle-mounted device receives the task instruction and performs corresponding operations.

[0086] The present application provides a task information identification device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the task information identification method in the above-mentioned embodiment 1.

[0087] Reference below Figure 4 , which shows a schematic diagram of the structure of a task information identification device suitable for implementing the embodiment of the present application. The task information identification device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The task information identification device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0088] like Figure 4As shown, the task information recognition device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the task information recognition device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the task information identification device to communicate with other devices wirelessly or by wire to exchange data. Although the task information identification device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have instead.

[0089] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0090] The task information recognition device provided by the present application adopts the task information recognition method in the above embodiment, which can solve the technical problem of how to improve the accuracy of speech recognition task extraction. Compared with the prior art, the beneficial effects of the task information recognition device provided by the present application are the same as the beneficial effects of the task information recognition method provided by the above embodiment, and the other technical features in the task information recognition device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0091] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0092] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0093] The present application provides a computer-readable medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the task information identification method in the above-mentioned embodiment.

[0094] The computer-readable medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0095] The computer-readable medium may be included in the task information identification device; or may exist independently without being incorporated into the task information identification device.

[0096] The computer-readable medium carries one or more programs, and when the one or more programs are executed by the task information identification device, the task information identification device can write computer program codes for performing the operations of the present application in one or more programming languages ​​or a combination thereof, and the programming languages ​​include object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0097] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0098] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0099] The readable medium provided by the present application is a computer-readable medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned task information recognition method, and can solve the technical problem of how to improve the accuracy of speech recognition task extraction. Compared with the prior art, the beneficial effects of the computer-readable medium provided by the present application are the same as the beneficial effects of the task information recognition method provided by the above-mentioned embodiment, and will not be repeated here.

[0100] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned task information identification method when executed by a processor.

[0101] The computer program product provided by this application can solve the technical problem of how to improve the accuracy of speech recognition task extraction. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the task information identification method provided by the above embodiment, which will not be repeated here.

[0102] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A task information identification method, characterized in that: The method comprises: Get voice information; Inputting the voice information into a preset deep learning model to obtain an intention recognition result, vehicle-mounted usage intention, and slot information, wherein the preset deep learning model at least includes a voice encoder, a predictor, and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer; Obtaining an expanded intention recognition result according to the vehicle-mounted use intention and the slot information combined with the intention recognition result; The task information is obtained according to the expanded intention recognition result.

2. The method according to claim 1, characterized in that The step of inputting a preset deep learning model according to the voice information to obtain an intention recognition result, a vehicle-mounted use intention, and slot information includes: Inputting the speech information into a preset deep learning model, converting the speech information through the speech encoder to obtain a feature vector; Inputting the feature vector into the predictor to obtain predicted text information; Inputting the predicted text information into the joint network layer for integration to generate an intent recognition result; An analysis is performed based on the intention recognition result to obtain vehicle usage intention and slot information.

3. The method according to claim 2, characterized in that The step of inputting the predicted text information into the joint network layer for integration to generate an intent recognition result includes: Inputting the predicted text information into the joint network layer; The predicted text information is integrated through the joint network layer to perform nonlinear changes and fusion operations to obtain the intent recognition result.

4. The method according to claim 1, characterized in that The step of obtaining an expanded intention recognition result according to the vehicle-mounted use intention and the slot information combined with the intention recognition result comprises: Expanding the vehicle usage intention and the slot information to obtain an extended text vocabulary, wherein the vehicle usage intention at least includes navigation, media control, air conditioning adjustment, and seat control, and the slot information at least includes a geographic location, a song name, and a contact; The extended text vocabulary is expanded to the intention recognition result to obtain an expanded intention recognition result.

5. The method according to claim 4, characterized in that Before the step of expanding the extended text vocabulary to the intention recognition result to obtain the extended intention recognition result, the method includes: Obtaining intent slot information according to the extended text vocabulary; Setting the intent slot information to a preset label; The step of expanding the extended text vocabulary to the intention recognition result to obtain an extended intention recognition result includes: When the preset tag is detected in the intention recognition result, the intention recognition result is expanded in combination with the extended text vocabulary to obtain an expanded intention recognition result.

6. The method according to claim 1, characterized in that The step of obtaining task information according to the expanded intention recognition result includes: Analyze the expanded intent recognition result to obtain the user intent and the slot information corresponding to the user intent; The task information is obtained by performing conversion according to the user intention and the slot information.

7. The method according to claim 1, characterized in that After the step of converting according to the user intention and the relevant slot information to obtain the task information, the method further includes: A task instruction is generated according to the task information and sent to the vehicle-mounted device, so that the vehicle-mounted device receives the task instruction and performs corresponding operations.

8. A task information identification device, characterized in that: The device comprises: An acquisition module, used for acquiring voice information; The acquisition module is further used to input the voice information into a preset deep learning model to obtain an intention recognition result, vehicle-mounted use intention and slot information, wherein the preset deep learning model at least includes a voice encoder, a predictor and a joint network layer, wherein the voice encoder and the predictor are both connected to the joint network layer; An expansion module, used to obtain an expanded intention recognition result according to the vehicle-mounted use intention and the slot information in combination with the intention recognition result; The analysis module is used to obtain task information according to the expanded intention recognition result.

9. A task information recognition device, characterized in that: The device comprises: a memory, a processor, and a task information identification program stored in the memory and executable on the processor, wherein the task information identification program is configured to implement the steps of the task information identification method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a task information identification program, and when the task information identification program is executed by the processor, the steps of the task information identification method according to any one of claims 1 to 7 are implemented.

Images