Vehicle-mounted voice interaction method and device, vehicle, medium and program

Abstract

The invention relates to the technical field of vehicles, in particular to a vehicle-mounted voice interaction method and device, a vehicle, a medium and a program, and the method comprises the steps: recognizing voice data inputted by a user; detecting a network state of the vehicle, and switching a processing link of the voice data according to the network state; and processing the voice data according to the processing link to generate a voice recognition result, and controlling the vehicle to execute the target operation according to the voice recognition result. Therefore, the problems that in the related technology, voice data processing is not intelligent and flexible enough in the aspect of end-cloud collaboration, the service strategy cannot be dynamically adjusted according to the real-time network and user requirements, and the overall user experience and service quality are affected are solved.

Description

Technical Field

[0001] The present application relates to the field of vehicle technology, and in particular to an in-vehicle voice interaction method, device, vehicle, medium and program. Background Art

[0002] With the rapid development of smart cars, in-vehicle voice assistants, as an important part of human-computer interaction, have gradually become one of the key technologies to improve driving experience and ensure driving safety. However, there are still many problems with in-vehicle voice assistants on the current market, especially in terms of network adaptability, command understanding and execution. These problems not only affect the user experience, but also limit the expansion and application of in-vehicle voice assistant functions to a certain extent.

[0003] In the related technologies, in-vehicle voice assistant solutions are mainly divided into two categories: cloud-based processing and services, and processing modes that completely rely on the vehicle side. Among them, cloud-side processing mainly utilizes powerful cloud computing resources to provide more complex and accurate voice recognition and natural language processing services, but once the network conditions are poor, it may cause service delays or unavailability, seriously affecting the user experience; while terminal-side processing is mainly independent of the network, can work without an Internet connection, and has a faster response speed, but due to the limited computing power and storage resources of vehicle-mounted equipment, the terminal-side model is usually simpler, resulting in limited voice recognition and natural language understanding capabilities, and unable to process complex user requests. Summary of the invention

[0004] The present application provides an in-vehicle voice interaction method, device, vehicle, medium and program to solve the problems in the related technology that voice data processing is not intelligent and flexible enough in end-cloud collaboration, and cannot dynamically adjust service strategies according to real-time network and user needs, affecting the overall user experience and service quality.

[0005] The first aspect of the present application provides an in-vehicle voice interaction method, comprising the following steps: identifying voice data input by a user; detecting the network status of the vehicle, and switching the processing link of the voice data according to the network status; processing the voice data according to the processing link to generate the voice recognition result, and controlling the vehicle to perform the target operation according to the voice recognition result.

[0006] Optionally, switching the processing link of the voice data according to the network status includes: if the network status is in a first target state, switching the processing link of the voice data to a cloud link; if the network status is in a second target state, switching the processing link of the voice data to a vehicle-side link.

[0007] Optionally, processing the voice data according to the processing link to generate the voice recognition result includes: if the processing link is a cloud link, using a voice recognition model and a collaborative model to process the voice data to generate a voice recognition result; if the processing link is a vehicle-side link, using a voice recognition algorithm and a natural language processing model to process the voice data to generate a voice recognition result.

[0008] Optionally, the use of a speech recognition model and a collaborative model to process the speech data to generate a speech recognition result includes: inputting the speech data into a speech recognition model, and the speech recognition model outputting corresponding speech text, wherein the speech recognition model extracts time series features from the speech data; matching a preset database according to the time series features to generate corresponding text content; inputting the text content into the collaborative model, and the collaborative model outputting a speech recognition result, wherein the collaborative model performs intent recognition and semantic analysis on the text content to generate corresponding user control intent, and generates a corresponding speech recognition result according to the user control intent.

[0009] Optionally, the processing of the speech data using a speech recognition algorithm and a natural language processing model to generate a speech recognition result includes: extracting key speech features of the speech data using a speech recognition algorithm to generate corresponding text data; inputting the text data into the natural language processing model, and the natural language processing model outputting a speech recognition result, wherein the natural language processing model performs semantic analysis and intent inference on the text data to generate a corresponding speech recognition result.

[0010] Optionally, the voice recognition result includes an instruction result and an inquiry result, wherein controlling the vehicle to perform a target operation based on the voice recognition result includes: if the voice recognition result is an instruction result, controlling the corresponding intelligent agent of the vehicle to execute the target instruction; if the voice recognition result is an inquiry result, controlling the vehicle to generate corresponding text data or voice data to feed back to the user.

[0011] The second aspect of the present application provides an in-vehicle voice interaction device, including: a recognition module for recognizing voice data input by a user; a detection module for detecting the network status of the vehicle and switching the processing link of the voice data according to the network status; a processing module for processing the voice data according to the processing link to generate the voice recognition result, and controlling the vehicle to perform a target operation according to the voice recognition result.

[0012] A third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform the in-vehicle voice interaction method as described in the above embodiment.

[0013] The fourth aspect of the present application provides a computer-readable storage medium on which a computer program is stored. The program is executed by a processor to perform the in-vehicle voice interaction method as described in the above embodiment.

[0014] The fifth aspect of the present application provides a computer program product, including a computer program or instructions, characterized in that when the computer program or instructions are executed, the in-vehicle voice interaction method as described in the above embodiment is implemented.

[0015] Therefore, this application has at least the following beneficial effects:

[0016] (1) The embodiments of the present application can recognize voice data input by the user; switch the processing link of the voice data according to the network status; process the voice data according to the processing link to generate a voice recognition result, and control the vehicle to perform the target operation according to the voice recognition result. By real-time monitoring of the network status and flexibly switching the cloud-side or terminal-side link, it ensures that the user's instructions can be responded to accurately and quickly, greatly improving the user experience, enhancing the overall performance and adaptability of the in-vehicle voice assistant, and enabling it to play a good role in different usage scenarios.

[0017] (2) The embodiments of the present application can use the speech recognition model and the collaborative model on the cloud to process the speech data to generate the speech recognition results, and use the speech recognition algorithm and the natural language processing model on the vehicle side to process the speech data to generate the speech recognition results, thereby enhancing the ability of the vehicle-mounted voice assistant to cope with complex network environments, greatly improving the accuracy and generalization ability of command processing, realizing efficient and accurate voice interaction, and meeting the user's diverse and natural command needs.

[0018] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The above and / or additional aspects and advantages of the present application are described below in conjunction with the accompanying drawings.

[0020] Figure 1 A flow chart of a vehicle-mounted voice interaction method provided according to an embodiment of the present application;

[0021] Figure 2 A schematic diagram of a method for constructing an in-vehicle voice assistant based on a large model according to an embodiment of the present application;

[0022] Figure 3 A schematic diagram of a process for enhanced generation of vehicle control agent retrieval according to an embodiment of the present application;

[0023] Figure 4 This is a block diagram of an in-vehicle voice interaction device provided according to an embodiment of the present application;

[0024] Figure 5 It is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0026] The following describes the in-vehicle voice interaction method, device, vehicle, storage medium and program of the embodiments of the present application with reference to the accompanying drawings.

[0027] Specifically, Figure 1 A flowchart of an in-vehicle voice interaction method provided in an embodiment of the present application.

[0028] like Figure 1 As shown, the vehicle-mounted voice interaction method includes the following steps:

[0029] In step S101 , voice data input by a user is recognized.

[0030] It is understandable that the embodiments of the present application can recognize the voice data input by the user to facilitate the subsequent analysis of the voice content.

[0031] In step S102, the network status of the vehicle is detected, and the processing link of the voice data is switched according to the network status.

[0032] It can be understood that the embodiments of the present application can detect the network status of the vehicle, switch the processing link of the voice data according to the network status, and ensure that user commands can be responded to accurately and quickly by monitoring the network status in real time and flexibly switching the cloud side or terminal side links, thereby greatly improving the user experience.

[0033] It should be noted that 1) Network status monitoring requires accurate network signal measurement tools and algorithms that can obtain signal strength (in dBm) and data transmission rate (in Mbps) in real time. The judgment threshold of a good network can be adjusted according to actual application requirements and network environment, and is generally set at a data transmission rate level that can ensure stable and efficient operation of cloud services.

[0034] In an embodiment of the present application, the processing link of voice data is switched according to the network status, including: if the network status is in a first target state, the processing link of voice data is switched to a cloud link; if the network status is in a second target state, the processing link of voice data is switched to a vehicle-side link.

[0035] The first target state may be a good network state, and the second target state may be a poor network state or a no network connection state.

[0036] It can be understood that the embodiment of the present application can switch the processing link of voice data to the cloud link when the network status is in the first target state, and switch the processing link of voice data to the vehicle-side link when the network status is in the second target state. By monitoring the network status in real time and flexibly switching the cloud-side or terminal-side links, it is ensured that user commands can be responded to accurately and quickly, greatly improving the user experience.

[0037] It should be noted that the network monitoring module in the vehicle-mounted equipment continuously collects and analyzes parameters such as the signal strength and data transmission rate of the network connection in real time; if the data transmission rate is higher than the set threshold, such as 5Mbps, the network is judged to be good; if it is lower than the threshold, the network is judged to be poor.

[0038] In step S103, the voice data is processed according to the processing link to generate a voice recognition result, and the vehicle is controlled to perform a target operation according to the voice recognition result.

[0039] It can be understood that the embodiments of the present application can generate voice recognition results according to the processing link by processing voice data, and control the vehicle to perform target operations according to the voice recognition results, thereby meeting the user's control needs, enhancing the overall performance and adaptability of vehicle voice recognition, enabling it to play a good role in different usage scenarios, and improving the user's experience.

[0040] In an embodiment of the present application, voice data is processed according to a processing link to generate a voice recognition result, including: if the processing link is a cloud link, a voice recognition model and a collaborative model are used to process the voice data to generate a voice recognition result; if the processing link is a vehicle-side link, a voice recognition algorithm and a natural language processing model are used to process the voice data to generate a voice recognition result.

[0041] It can be understood that the embodiments of the present application can use the speech recognition model and the collaborative model to process the voice data to generate a speech recognition result when the processing link is a cloud link, and use the speech recognition algorithm and the natural language processing model to process the voice data to generate a speech recognition result when the processing link is a vehicle-side link, so as to accurately and flexibly process the voice data to meet the user's diverse and natural command requirements and enhance the user experience.

[0042] In an embodiment of the present application, a speech recognition model and a collaborative model are used to process speech data to generate a speech recognition result, including: inputting speech data into the speech recognition model, and the speech recognition model outputs the corresponding speech text, wherein the speech recognition model extracts time series features in the speech data; generates corresponding text content according to the time series feature matching preset database; inputting the text content into the collaborative model, and the collaborative model outputs the speech recognition result, wherein the collaborative model performs intent recognition and semantic analysis on the text content to generate the corresponding user control intent, and generates the corresponding speech recognition result according to the user control intent.

[0043] Among them, the preset database can be selected according to actual needs without specific limitation.

[0044] It can be understood that the embodiments of the present application can input voice data into a voice recognition model, and the voice recognition model outputs corresponding voice text, wherein the voice recognition model extracts time series features in the voice data; generates corresponding text content based on the time series feature matching of a preset database; and inputs the text content into a collaborative model, and the collaborative model outputs a voice recognition result, wherein the collaborative model performs intent recognition and semantic analysis on the text content to generate corresponding user control intent, and generates corresponding voice recognition results based on the user control intent to improve the accuracy of voice recognition in the cloud link.

[0045] It should be noted that cloud-side speech recognition: using the powerful computing power of cloud servers and advanced speech recognition algorithms to accurately convert user input speech into text. Collaborative model processing: based on deep learning collaborative models, combined with a large amount of text data and contextual information, accurately judge user intent, and enter the domain to call other intelligent agents according to preset rules and algorithms.

[0046] Specifically, the cloud side link processes:

[0047] (1) Cloud-side speech recognition: When the user issues a voice command, the vehicle-mounted device quickly transmits the voice data to the cloud server. The cloud server uses its powerful computing resources and advanced speech recognition models, such as neural network models based on deep learning, to extract features and match patterns of the speech, and accurately convert it into clear text information.

[0048] (2) Collaborative model processing: After receiving the speech-to-text results, the collaborative model combines the contextual information obtained from the vehicle-mounted device, such as the previous conversation content and user preferences, to accurately determine the user's intention through deep analysis and semantic understanding. Then, according to the preset rules and agent calling strategy, the task is assigned to the corresponding agent for processing, such as the vehicle control agent.

[0049] In an embodiment of the present application, speech data is processed using a speech recognition algorithm and a natural language processing model to generate a speech recognition result, including: using a speech recognition algorithm to extract key speech features of the speech data to generate corresponding text data; inputting the text data into a natural language processing model, and the natural language processing model outputting a speech recognition result, wherein the natural language processing model performs semantic analysis and intent inference on the text data to generate a corresponding speech recognition result.

[0050] It can be understood that the embodiments of the present application can use a speech recognition algorithm to extract key speech features of speech data to generate corresponding text data; the text data is input into a natural language processing model, and the natural language processing model outputs a speech recognition result, wherein the natural language processing model performs semantic analysis and intent inference on the text data to generate a corresponding speech recognition result, which can not only accurately understand the user's instructions, but also provide more intelligent and personalized services, greatly enhancing the user experience and system reliability.

[0051] It should be noted that (1) end-side speech recognition: with the help of the local speech recognition chip and optimization algorithm of the vehicle-mounted device, the speech is recognized and converted into text. In the case of poor network, the local speech recognition module of the vehicle-mounted device starts working. This module uses the optimized speech recognition algorithm to perform noise reduction, feature extraction and other processing on the collected speech, and converts the speech into text.

[0052] (2) Natural language understanding on the device: The local pre-loaded natural language processing model is used to analyze and understand the text. The natural language understanding model pre-loaded on the device performs semantic parsing and intent inference on the generated text. This model is compressed and optimized to adapt to the limited computing resources of the vehicle-mounted equipment and can understand common instructions and user needs.

[0053] Specifically, if Figure 3 As shown in the figure, vehicle control and ecological application instructions are completed using the concept of intelligent agents, abandoning the traditional slot-based keyword matching method, using a large model to deeply understand user instructions, generating voice instructions that meet the requirements of the vehicle-side system interface, and using function calling technology, that is, the process of using user voice commands to call specific functions or services, to achieve efficient and accurate interaction with the vehicle-side system interface and complete the execution of instructions.

[0054] Step 1: User makes an inquiry

[0055] Step 2: Intent understanding: Intent understanding means receiving a user query, understanding the user's query intent, breaking it down into several selection modules, and using traditional natural language classification models to make judgments.

[0056] Step 3: Query rewriting: retrieval fusion, similarity construction based on the original query, each query is retrieved using a different retrieval method, and reverse sorting is performed based on the retrieval content as the model input

[0057] Step 4: Multi-way search:

[0058] Keyword search: Segment the query and search the knowledge base based on the segmented content

[0059] Semantic search: vectorize the query and search the query content

[0060] Online query: query related content based on the network

[0061] Step 5: Rearrange: Sort the content in the multi-query in step 4 in descending order and select the content with strong relevance

[0062] Step 6: Prompt word or guide word synthesis: Create prompt words or guide words based on the rearranged content

[0063] Step 7: Input to the big model, and the big model generates content.

[0064] In an embodiment of the present application, the voice recognition result includes an instruction result and an inquiry result, wherein the vehicle is controlled to perform a target operation according to the voice recognition result, including: if the voice recognition result is an instruction result, the corresponding intelligent agent of the vehicle is controlled to execute the target instruction; if the voice recognition result is an inquiry result, the vehicle is controlled to generate corresponding text data or voice data and feedback it to the user.

[0065] It can be understood that in the embodiment of the present application, when the voice recognition result is a command result, the corresponding intelligent agent of the vehicle is controlled to execute the target command; when the voice recognition result is an inquiry result, the vehicle is controlled to generate corresponding text data or voice data and feedback it to the user, so as to meet the user's diverse needs and enhance the user experience.

[0066] It should be noted that the voice recognition results of the present application may include adjusting the temperature in the car, switching radio channels, etc., without specific limitation.

[0067] According to the in-vehicle voice interaction method proposed in the embodiment of the present application, the voice data input by the user is recognized; the processing link of the voice data is switched according to the network status; the voice data is processed according to the processing link to generate a voice recognition result, and the vehicle is controlled to perform the target operation according to the voice recognition result. By real-time monitoring of the network status and flexibly switching the cloud-side or terminal-side links, it is ensured that the user's instructions can be responded to accurately and quickly, which greatly improves the user experience and enhances the overall performance and adaptability of the in-vehicle voice assistant, so that it can play a good role in different usage scenarios.

[0068] The in-vehicle voice interaction method of the present application will be described in detail below in conjunction with specific embodiments, wherein the in-vehicle device includes a microphone for collecting voice, a processor for running a local voice recognition and natural language understanding module, a network module for communicating with a cloud server, and a storage unit for storing local models and data; the cloud server has a powerful computing cluster, a large-capacity storage device, and a high-speed network connection to support cloud-side voice recognition and collaborative model processing, as follows:

[0069] Embodiment 1:

[0070] Voice interaction processing based on cloud links:

[0071] (1) Voice data collection and preprocessing

[0072] Microphone collection: Users issue voice commands through the microphone on the vehicle-mounted device.

[0073] Preprocessing: The system performs preliminary processing on the collected voice signals, such as noise reduction and gain adjustment, to improve the quality of subsequent processing.

[0074] (2) Upload to the cloud server

[0075] Data transmission: The pre-processed voice data is quickly uploaded to the cloud server, ensuring that it can be processed using powerful computing resources under good network conditions.

[0076] (3) Speech recognition model processing

[0077] Input speech data: Input speech data into the pre-trained speech recognition model.

[0078] Time series feature extraction:

[0079] Feature extraction: The speech recognition model first extracts time series features from the speech data. These features may include phonemes, Mel-spectrogram coefficients (MFCCs), etc., which can capture the time-varying characteristics of speech.

[0080] Feature vectorization: Convert the extracted time series features into a vector form suitable for model processing.

[0081] Matching preset database:

[0082] Text generation: Based on the extracted time series features, the speech recognition model searches for the most similar pattern in the internal preset speech-text mapping database and generates the corresponding text content.

[0083] Text correction: Use language models to further optimize the generated text, correct possible grammatical or semantic errors, and ensure the accuracy of the text.

[0084] (4) Collaborative model processing

[0085] Text content input: The generated text content is fed into the collaborative model.

[0086] Intent recognition and semantic analysis:

[0087] Contextual understanding: The collaborative model combines contextual information (such as previous conversation records, user preferences, etc.) to accurately determine the user's intention through deep analysis and semantic understanding.

[0088] Intent classification: According to the user's intention, the task is assigned to the corresponding agent for processing. For example, the vehicle control agent is responsible for vehicle control related instructions.

[0089] User control intention generation: The collaborative model outputs the final user control intention, which may be a specific command (such as "set the air conditioning temperature to 22 degrees") or a query request (such as "the location of nearby gas stations").

[0090] (5) Generate speech recognition results

[0091] Instruction results vs. inquiry results:

[0092] If it is an instruction result, the collaborative model will generate a clear operation instruction and pass it to the corresponding agent for execution.

[0093] If it is a query result, text or voice feedback is generated and prepared to be returned to the user.

[0094] (6) Control the vehicle to perform the target operation

[0095] Calling the agent: For the instruction results, the system calls the vehicle control agent or other related modules to execute specific commands.

[0096] Feedback to users: Regardless of whether it is the result of a command or an inquiry, the system will convert the result into voice feedback to the user through speech synthesis technology, or directly display it on the in-vehicle screen.

[0097] (7) Exception handling

[0098] Error recovery mechanism: If a failure occurs during cloud processing, such as a network interruption, the system will immediately switch back to the vehicle-side link to continue processing unfinished tasks, ensuring that the user experience is not affected.

[0099] User prompts: For instructions that cannot be recognized or executed, the system should provide clear feedback to the user, explaining the reasons and guiding the next steps.

[0100] For example, suppose the user says, "Set the air conditioning temperature to 22 degrees." The user speaks the command through the microphone, and the system reduces noise and adjusts the gain of the voice signal; the preprocessed voice data is uploaded to the cloud server; the speech recognition model processes: based on the time series characteristics, it searches the preset database for the most similar pattern and generates the text "Set the air conditioning temperature to 22 degrees"; the collaborative model combines the context information to understand that this is a control command intended to set the air conditioning temperature; it outputs the user's control intention, that is, the command "Set the air conditioning temperature to 22 degrees"; the system calls the vehicle control agent to execute the operation of setting the air conditioning temperature, and informs the user through speech synthesis technology that "the air conditioning temperature has been set to 22 degrees."

[0101] Embodiment 2:

[0102] Voice interaction processing based on vehicle-side link:

[0103] (1) Voice data collection and preprocessing

[0104] Microphone collection: Users issue voice commands through the microphone on the vehicle-mounted device.

[0105] Preprocessing: The system performs preliminary processing on the collected voice signals, such as noise reduction and gain adjustment, to improve the quality of subsequent processing.

[0106] (2) Speech recognition algorithm processing

[0107] Extract key speech features:

[0108] Feature extraction: Use optimized speech recognition algorithms to extract key features of speech data, such as phonemes, Mel-spectral coefficients (MFCCs), prosodic features, etc. These features can capture the temporal variation of speech and help improve recognition accuracy.

[0109] Feature vectorization: Convert the extracted key features into a vector form suitable for algorithm processing to facilitate subsequent analysis and matching.

[0110] Generate text data:

[0111] Pattern matching: Based on the extracted key speech features, the most similar pattern is searched in the local preset speech-text mapping database, and the corresponding text content is generated.

[0112] Text correction: Use the local language model to further optimize the generated text, correct possible grammatical or semantic errors, and ensure the accuracy of the text.

[0113] (3) Natural Language Processing Model Processing

[0114] Text data input: The generated text data is fed into the natural language processing model pre-loaded on the onboard device.

[0115] Semantic parsing and intent inference:

[0116] Contextual understanding: The NLP model combines contextual information (such as previous conversation records, user preferences, etc.) to accurately determine the user's intention through deep analysis and semantic understanding.

[0117] Intent classification: According to the user's intention, the task is assigned to the corresponding agent for processing. For example, the vehicle control agent is responsible for vehicle control related instructions.

[0118] User control intent generation: The NLP model outputs the final user control intent, which may be a specific command (such as "set the air conditioning temperature to 22 degrees") or a query request (such as "the location of nearby gas stations").

[0119] (4) Generate speech recognition results

[0120] Instruction results vs. inquiry results:

[0121] If it is an instruction result, the NLP model will generate a clear operation instruction and pass it to the corresponding agent for execution.

[0122] If it is a query result, text or voice feedback is generated and prepared to be returned to the user.

[0123] (5) Control the vehicle to perform the target operation

[0124] Calling the agent: For the instruction results, the system calls the vehicle control agent or other related modules to execute specific commands.

[0125] Feedback to users: Regardless of whether it is the result of a command or an inquiry, the system will convert the result into voice feedback to the user through speech synthesis technology, or directly display it on the in-vehicle screen.

[0126] (6) Exception handling

[0127] Error recovery mechanism: If a problem is encountered during processing (such as an unrecognized instruction), the system will try to provide help or suggest that the user rephrase the instruction.

[0128] User prompts: For instructions that cannot be recognized or executed, the system should provide clear feedback to the user, explaining the reasons and guiding the next steps.

[0129] In summary, this application is based on speech recognition technology, natural language processing technology and deep learning principles in artificial intelligence; speech recognition converts speech into processable text form by extracting features and matching patterns of speech signals; natural language understanding uses deep learning models to perform semantic analysis and intent inference on text, and by learning from large amounts of text data, the model can automatically extract regularities and features in the language to understand user needs; the neural network model in deep learning can automatically learn hidden patterns and features in the data through multi-layer neuron structures and complex connections, thereby achieving accurate speech recognition and natural language understanding; Function calling technology is based on predefined interface specifications and calling rules to achieve efficient communication and collaboration between different modules.

[0130] Next, the in-vehicle voice interaction device proposed according to the embodiment of the present application is described with reference to the accompanying drawings.

[0131] Figure 4 It is a block diagram of the in-vehicle voice interaction device of an embodiment of the present application.

[0132] like Figure 4 As shown, the vehicle-mounted voice interaction device 10 includes: a recognition module 100, a detection module 200 and a processing module 300.

[0133] Among them, the recognition module 100 is used to recognize the voice data input by the user; the detection module 200 is used to detect the network status of the vehicle and switch the processing link of the voice data according to the network status; the processing module 300 is used to process the voice data according to the processing link to generate a voice recognition result, and control the vehicle to perform the target operation according to the voice recognition result.

[0134] It should be noted that the aforementioned explanation of the embodiment of the in-vehicle voice interaction method is also applicable to the in-vehicle voice interaction device of this embodiment, and will not be repeated here.

[0135] According to the in-vehicle voice interaction device proposed in the embodiment of the present application, the voice data input by the user is recognized; the processing link of the voice data is switched according to the network status; the voice data is processed according to the processing link to generate a voice recognition result, and the vehicle is controlled to perform the target operation according to the voice recognition result. By real-time monitoring of the network status and flexibly switching the cloud-side or terminal-side link, it is ensured that the user's instructions can be responded to accurately and quickly, which greatly improves the user experience and enhances the overall performance and adaptability of the in-vehicle voice assistant, so that it can play a good role in different usage scenarios.

[0136] Figure 5 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0137] A memory 501 , a processor 502 , and a computer program stored in the memory 501 and executable on the processor 502 .

[0138] When the processor 502 executes the program, the in-vehicle voice interaction method provided in the above embodiment is implemented.

[0139] Furthermore, the vehicle also includes:

[0140] The communication interface 503 is used for communication between the memory 501 and the processor 502 .

[0141] The memory 501 is used to store computer programs that can be executed on the processor 502 .

[0142] The memory 501 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0143] If the memory 501, the processor 502 and the communication interface 503 are implemented independently, the communication interface 503, the memory 501 and the processor 502 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0144] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can communicate with each other through an internal interface.

[0145] The processor 502 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0146] An embodiment of the present application also provides a computer-readable storage medium on which a computer program or instruction is stored. When the computer program or instruction is executed by a processor, the above-mentioned in-vehicle voice interaction method is implemented.

[0147] An embodiment of the present application also provides a computer program product, including a computer program or instructions, characterized in that when the computer program or instructions are executed, the above-mentioned in-vehicle voice interaction method is implemented.

[0148] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0149] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0150] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0151] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of multiple of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0152] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

Claims

1. A vehicle-mounted voice interaction method, characterized in that: The following steps are involved: Recognize voice data input by the user; Detecting a network status of the vehicle, and switching a processing link of the voice data according to the network status; The voice data is processed according to the processing link to generate the voice recognition result, and the vehicle is controlled to perform a target operation according to the voice recognition result.

2. The vehicle-mounted voice interaction method according to claim 1, characterized in that: The switching of the processing link of the voice data according to the network state includes: If the network state is in the first target state, switching the processing link of the voice data to a cloud link; If the network state is in the second target state, the processing link of the voice data is switched to the vehicle-side link.

3. The vehicle-mounted voice interaction method according to claim 1, characterized in that: The processing of the voice data according to the processing link to generate the voice recognition result includes: If the processing link is a cloud link, the speech data is processed using a speech recognition model and a collaborative model to generate a speech recognition result; If the processing link is a vehicle-side link, the voice data is processed using a voice recognition algorithm and a natural language processing model to generate a voice recognition result.

4. The vehicle-mounted voice interaction method according to claim 3, characterized in that: The using the speech recognition model and the collaborative model to process the speech data to generate a speech recognition result includes: The speech data is input into a speech recognition model, and the speech recognition model outputs corresponding speech text, wherein the speech recognition model extracts time series features from the speech data; and generates corresponding text content according to the time series features matched with a preset database; The text content is input into the collaborative model, and the collaborative model outputs a speech recognition result, wherein the collaborative model performs intent recognition and semantic analysis on the text content to generate a corresponding user control intent, and generates a corresponding speech recognition result according to the user control intent.

5. The vehicle-mounted voice interaction method according to claim 3, characterized in that: The step of processing the speech data using a speech recognition algorithm and a natural language processing model to generate a speech recognition result includes: Extracting key speech features of the speech data using a speech recognition algorithm to generate corresponding text data; The text data is input into the natural language processing model, and the natural language processing model outputs a speech recognition result, wherein the natural language processing model performs semantic analysis and intent inference on the text data to generate a corresponding speech recognition result.

6. The vehicle-mounted voice interaction method according to claim 1, characterized in that: The speech recognition result includes a command result and an inquiry result, wherein controlling the vehicle to perform a target operation according to the speech recognition result includes: If the speech recognition result is a command result, the intelligent agent corresponding to the vehicle is controlled to execute the target command; If the voice recognition result is an inquiry result, the vehicle is controlled to generate corresponding text data or voice data and feed it back to the user.

7. A vehicle-mounted voice interaction device, characterized in that: include: A recognition module, used to recognize voice data input by a user; A detection module, used to detect the network status of the vehicle and switch the processing link of the voice data according to the network status; A processing module is used to process the voice data according to the processing link to generate the voice recognition result, and control the vehicle to perform a target operation according to the voice recognition result.

8. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the in-vehicle voice interaction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by the processor, it is used to implement the in-vehicle voice interaction method as described in any one of claims 1-6.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed, the in-vehicle voice interaction method as described in any one of claims 1-6 is implemented.

Images