Voice instruction processing method, device and equipment and computer storage medium

Abstract

The invention discloses a voice instruction processing method and device, equipment and a computer storage medium. The method comprises the following steps: receiving a first voice instruction; sending the first voice instruction and a user-defined instruction stored in the vehicle-mounted equipment to a server, wherein an audio recognition model is deployed in the server; receiving a matching degree between the first voice instruction and a user-defined instruction, wherein the matching degree is generated based on an audio recognition model; when it is determined that the matching degree is larger than or equal to a credibility threshold value corresponding to the user-defined instruction, a first operation instruction corresponding to the user-defined instruction is determined according to a first corresponding relation, and the first corresponding relation represents the corresponding relation between the user-defined instruction and one or more operation instructions; and executing the first operation instruction. According to the embodiment of the invention, the vehicle-mounted equipment can respond to the user-defined voice instruction, and the driving experience of the intelligent vehicle is improved.

Description

Technical Field

[0001] The present application belongs to the field of intelligent automobile technology, and in particular, relates to a voice command processing method, device, equipment and computer storage medium. Background Art

[0002] With the rapid development of smart cars and artificial intelligence, users have higher and higher requirements for the convenience of driving smart cars. Users hope to be able to safely and conveniently realize various functions of smart cars while driving smart cars.

[0003] In related technologies, smart cars can receive and recognize user voice commands to obtain operation instructions corresponding to the user's voice commands, so as to control the vehicle to execute the operation instructions and respond to user needs.

[0004] However, in the related art, the command words corresponding to the voice commands are usually prepared in advance by relevant technicians. The vehicle will only execute the corresponding operation command after recognizing the command words, which is inconsistent with the development needs of personalized and convenient smart cars. Summary of the invention

[0005] The embodiments of the present application provide a voice command processing method, apparatus, device and computer storage medium, which can enable an in-vehicle device to respond to user-defined voice commands and enhance the driving experience of a smart car.

[0006] In a first aspect, an embodiment of the present application provides a voice command processing method, which is applied to an in-vehicle device, and the method includes: receiving a first voice command; sending the first voice command and a user-defined command stored in the in-vehicle device to a server, in which an audio recognition model is deployed; receiving a matching degree between the first voice command and the user-defined command, wherein the matching degree is generated based on the audio recognition model; when it is determined that the matching degree is greater than or equal to a credibility threshold corresponding to the user-defined command, determining a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, wherein the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions; and executing the first operation instruction.

[0007] In one embodiment, after sending the first voice instruction and the user-defined instruction to the server, the method also includes: receiving text information corresponding to the first voice instruction, the text information being generated based on an audio recognition model; when it is determined that the degree of match is less than a credibility threshold corresponding to the customized instruction, determining a second operation instruction corresponding to the text information based on the text information; and executing the second operation instruction.

[0008] In one embodiment, before receiving the first voice instruction, the method also includes: sending the user-defined instruction to the server so that the server determines the credibility threshold corresponding to the user-defined instruction based on the audio recognition model; and receiving the credibility threshold corresponding to the user-defined instruction.

[0009] In the second aspect, an embodiment of the present application provides a voice command processing method, which is applied to a server, and the method includes: receiving a first voice command and a user-defined command sent by a vehicle-mounted device; calculating the matching degree of the first voice command and the user-defined command through an audio recognition model to obtain the matching degree of the first voice command and the user-defined command; sending the credibility threshold and the matching degree to the vehicle-mounted device, so that the vehicle-mounted device determines the first operation instruction corresponding to the user-defined instruction according to the first corresponding relationship when determining that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, and executes the first operation instruction.

[0010] In one implementation, the method further includes: receiving a user-defined instruction; and determining a credibility threshold corresponding to the user-defined instruction through an audio recognition model.

[0011] In one implementation, the credibility threshold corresponding to the user-defined instruction is determined through an audio recognition model, including: determining a pinyin sequence corresponding to the user-defined instruction; analyzing the pinyin sequence through the audio recognition model to determine the credibility threshold corresponding to the pinyin sequence.

[0012] In one embodiment, after receiving a first voice command sent by a vehicle and a user-defined command, the method further includes: converting the first voice command into text through an audio recognition model to obtain text information corresponding to the first voice command; sending the text information corresponding to the first voice command to a vehicle-mounted device, so that the vehicle-mounted device determines a second operation instruction corresponding to the text information based on the text information when determining that the degree of match is less than a credibility threshold, and executes the second operation instruction.

[0013] In one embodiment, sending text information corresponding to a first voice command to a vehicle-mounted device includes: determining a credibility score corresponding to the text information through an audio recognition model; and when it is determined that the credibility score is greater than or equal to a preset credibility threshold, sending the text information corresponding to the first voice command to the vehicle-mounted device.

[0014] In a third aspect, an embodiment of the present application provides a voice command processing device, which is applied to an in-vehicle device, and the device includes:

[0015] A first receiving module, used to receive a first voice command;

[0016] A sending module, used for sending the first voice command and the user-defined command stored in the vehicle-mounted device to the server, where the audio recognition model is deployed;

[0017] A second receiving module is used to receive a matching degree between the first voice command and the user-defined command, where the matching degree is generated based on the audio recognition model;

[0018] A determination module, configured to determine, when it is determined that the matching degree is greater than or equal to a credibility threshold corresponding to the user-defined instruction, a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, wherein the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions;

[0019] The execution module is used to execute the first operation instruction.

[0020] In a fourth aspect, an embodiment of the present application provides a voice command processing device, which is applied to a server, and the device includes:

[0021] A receiving module, used for receiving a first voice command sent by the vehicle-mounted device;

[0022] A first determination module, configured to calculate a matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain a matching degree between the first voice instruction and the user-defined instruction;

[0023] The sending module is used to send the matching degree to the vehicle-mounted device, so that when the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, it determines the first operation instruction corresponding to the user-defined instruction according to the first corresponding relationship and executes the first operation instruction.

[0024] In a fifth aspect, an embodiment of the present application provides a voice command processing device, the device comprising: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the voice command processing method in the first aspect or any one of the embodiments of the first aspect; or, it implements the voice command processing method in the second aspect or any one of the embodiments of the second aspect.

[0025] In the sixth aspect, an embodiment of the present application provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the voice instruction processing method in the first aspect or any one of the embodiments of the first aspect is implemented; or, the voice instruction processing method in the second aspect or any one of the embodiments of the second aspect is implemented.

[0026] In the seventh aspect, an embodiment of the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes a voice instruction processing method as in the first aspect or any one of the embodiments of the first aspect; or, implements the voice instruction processing method as in the second aspect or any one of the embodiments of the second aspect.

[0027] The voice command processing method, device, equipment and computer storage medium of the embodiment of the present application can transmit the first voice command and the user-defined command pre-stored in the vehicle-mounted device to the server after the vehicle-mounted device receives the first voice command. After receiving the first voice command and the user-defined command, the server can use the internally deployed audio recognition model to determine the matching degree between the first voice command and the user-defined command transmission, and feed back the matching degree to the vehicle-mounted device, and the vehicle-mounted device makes a decision based on the credibility threshold and matching degree corresponding to the user-defined command, that is, when it is determined that the matching degree is greater than or equal to the credibility threshold, the first operation instruction corresponding to the user-defined instruction is determined and executed according to the first corresponding relationship. It can be understood that in the embodiment of the present application, through the user-defined command set by the user on the vehicle-mounted device end, each time after receiving the first voice command, the corresponding vehicle-mounted device uploads, receives and arbitrates, so that different vehicle-mounted devices can execute corresponding operation instructions according to the user-defined commands set by different users, and the different vehicle-mounted devices are affected by each other, so as to realize user-level custom command control. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0029] Figure 1 A flow chart of a method for processing voice commands provided by an embodiment of the present application is shown;

[0030] Figure 2 A flow chart of a voice command processing method provided by another embodiment of the present application is shown;

[0031] Figure 3 A schematic diagram of a process of sending a text message provided by an embodiment of the present application is shown;

[0032] Figure 4 A flow chart of a voice command processing method provided by another embodiment of the present application is shown;

[0033] Figure 5 A schematic diagram of a process for determining a credibility threshold provided by an embodiment of the present application is shown;

[0034] Figure 6 is a structural diagram of a voice command processing device provided by another embodiment of the present application;

[0035] Figure 7 is a structural diagram of a voice command processing device provided by another embodiment of the present application;

[0036] Figure 8 It is a structural diagram of a voice command processing device provided in yet another embodiment of the present application. DETAILED DESCRIPTION

[0037] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0038] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0039] With the development of smart car technology and artificial intelligence technology, the functions of smart cars are gradually enriched. In order to ensure that users can use various functions of the vehicle safely and conveniently while driving the vehicle, the relevant technology can obtain the user's voice command and analyze the user's voice command to determine the prompt word used in the voice command, and then according to the determined prompt word, the vehicle performs the corresponding function.

[0040] However, the prompt words in the related art are usually pre-formulated by technicians, and these prompt words are for all users, that is, the prompts for the voice commands corresponding to all users are the same. However, different users may have different habits and needs, and only the preset prompt words cannot provide users with a convenient and comfortable use experience. For example, a user may be accustomed to opening the car window and the audio playback device at the same time. At this time, the user may need to enter two or even more voice commands, causing inconvenience to the user. Or, the user hopes to be able to personalize the voice commands corresponding to the vehicle functions, but the related art cannot respond to the user's personalized voice commands.

[0041] In order to solve the problems of the prior art, the embodiments of the present application provide a method, device, equipment and computer storage medium for processing voice commands. The following first introduces the method for processing voice commands provided in the embodiments of the present application.

[0042] Figure 1 FIG. 1 is a flow chart showing a method for processing voice commands according to an embodiment of the present application. Figure 1 As shown, the voice command processing method includes the following steps:

[0043] S110: The vehicle-mounted device receives a first voice command.

[0044] S120: The vehicle-mounted device sends the first voice command and the user-defined command stored in the vehicle-mounted device to the server.

[0045] S130: The server receives a first voice command and a user-defined command sent by the vehicle-mounted device.

[0046] S140. The server calculates a matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain a matching degree between the first voice instruction and the user-defined instruction.

[0047] S150. The server sends the matching degree to the vehicle-mounted device.

[0048] S160: The in-vehicle device receives a matching degree between the first voice command and the user-defined command, where the matching degree is generated based on an audio recognition model.

[0049] S170. When the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, the vehicle-mounted device determines a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, where the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions.

[0050] S180: The vehicle-mounted device executes the first operation instruction.

[0051] In some embodiments, in S110, the first voice command may be a voice command of a user received by the vehicle-mounted device. In one example, the first voice command of the user may be received via a vehicle-mounted microphone.

[0052] Exemplarily, the user-defined instruction may be an instruction set by the user according to his / her own needs. The vehicle-mounted device may execute one or more vehicle functions according to the user-defined instruction.

[0053] Exemplarily, the user-defined instruction may be a text instruction or the user-defined instruction may be an audio instruction.

[0054] In one example, a user may enter custom instruction text into an in-vehicle device.

[0055] In some embodiments, in S120, an audio recognition model may be deployed in the server.

[0056] Exemplarily, the in-vehicle device may establish a connection with the server. After the in-vehicle device establishes a connection with the server, the in-vehicle device may send the first voice command and the user's corresponding custom command to the server.

[0057] In one example, the vehicle-mounted device may establish a communication connection with the server. In another example, the vehicle-mounted device may establish an electrical connection with the server.

[0058] In some embodiments, in S130, after receiving the first voice command and the user-defined command sent by the vehicle-mounted device, the server can execute step S140, that is, the first voice command and the user-defined command voice command recognition model can be used to enable the audio recognition model to calculate the matching degree of the first voice command and the user-defined command, thereby obtaining the matching degree between the first voice command and the user-defined command.

[0059] In one example, the audio recognition model can be initialized first, that is, a pre-trained audio recognition model is used. This model is usually built based on a deep neural network, such as a deep neural network-hidden Markov model (DNN-HMM), a long short-term memory network (LSTM) or a Transformer architecture model.

[0060] Further, after the model receives the first voice instruction and the user-defined instruction, the first voice instruction and the user-defined instruction can be preprocessed. Among them, the preprocessing of the first voice instruction can include operations such as sampling, quantization and framing. For example, the audio signal is discretized according to a certain sampling rate (for example, 16kHz), and the length of each frame of audio can be set to about 20-30 milliseconds, and there is a partial overlap between adjacent frames (such as 10-15 milliseconds of overlap). At the same time, the audio may be normalized so that the amplitude range of the audio signal meets the input requirements of the model. In addition, after preprocessing the first voice instruction, acoustic features can be extracted from each frame of audio. Among them, the acoustic features can be extracted by Mel frequency cepstrum coefficients. The extraction process is to first perform a fast Fourier transform (FFT) on the audio frame, convert the time domain signal into a frequency domain signal, and then filter the frequency domain signal through a Mel filter bank, and finally take the logarithm of the filtered result and perform a discrete cosine transform (DCT) to obtain the MFCC feature vector. These feature vectors are used as input to the speech recognition model. Furthermore, the extracted acoustic features can be used to identify a voice command model, and the model outputs a text sequence corresponding to the audio.

[0061] A rule-based pinyin conversion tool or a neural network model can be used. For example, the conversion can be performed through a Chinese character-pinyin mapping table, and the pronunciation of polyphonetic characters can be determined according to simple rules (such as word frequency, part of speech, etc.). The neural network model is trained on a large amount of Chinese character-pinyin annotation data to learn the complex mapping relationship from Chinese characters to pinyin. For the output text sequence, word segmentation may be required, especially for languages ​​such as Chinese that do not have obvious word boundaries. A statistical-based word segmentation algorithm (such as hidden Markov model word segmentation) or a rule-based word segmentation algorithm (such as maximum matching method) can be used, and each word or character can be converted into pinyin. Such a conversion is completed for both the first voice instruction and the user-defined instruction to obtain the corresponding pinyin sequence.

[0062] Furthermore, features can be extracted from the Pinyin sequence. Features such as the syllable structure of Pinyin (including initials, finals, and tones), the length of Pinyin, and the frequency of Pinyin combinations can be considered. A suitable distance measurement method is used to calculate the distance between the Pinyin features corresponding to the first voice instruction and the user-defined instruction. For example, the edit distance (LevenshteinDistance) can be used, which measures the minimum number of editing operations (insertion, deletion, replacement) required to convert one Pinyin sequence into another Pinyin sequence. Assume that the first voice instruction Pinyin sequence is A = [a1, a2, ..., an], and the user-defined audio Pinyin sequence is B = [b1, b2, ..., bn], and the edit distance is calculated by a dynamic programming algorithm.

[0063] Finally, the similarity is calculated based on the distance measurement result. For example, the similarity can be defined as the following formula (1):

[0064]

[0065] Among them, D is the calculated distance, that is, the edit distance, D max The maximum possible distance can be determined based on factors such as the length of the pinyin sequence. The similarity is then normalized so that the match value is between 0 and 1, where 0 represents a complete mismatch and 1 represents a complete match.

[0066] In some embodiments, in S150, the server sends the matching degree to the vehicle-mounted device. Further, in S160, the vehicle-mounted device receives the matching degree between the first voice command and the user-defined command.

[0067] In some embodiments, in S170, when the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, the vehicle-mounted device determines the first operation instruction corresponding to the user-defined instruction based on the first corresponding relationship, and the first corresponding relationship represents the correspondence between the user-defined instruction and one or more operation instructions.

[0068] Exemplarily, the credibility threshold corresponding to the user-defined instruction may be a credibility threshold pre-stored in the vehicle-mounted device. Moreover, the user voice instruction has a corresponding relationship with its credibility threshold.

[0069] Exemplarily, after the vehicle-mounted device receives the matching degree, it is determined whether the matching degree is greater than the credibility threshold. When it is determined that the matching degree is greater than or equal to the credibility threshold, the first operation instruction corresponding to the user-defined instruction can be determined according to the first corresponding relationship. Among them, the vehicle-mounted device can implement different functions of the vehicle according to the first operation instruction. For example, the first instruction can be to open the window; or, the first operation instruction can also be to turn on the audio playback device. It can be understood that the first operation instruction is not limited in the embodiment of the present application.

[0070] Exemplarily, the first corresponding relationship may be set by the user according to his / her own needs and system personalization.

[0071] Exemplarily, first, the user can set the name corresponding to the custom instruction, and further, can configure the first operation instruction corresponding to the custom instruction. The custom instruction set by the user can be one or more. And each user-defined instruction can be configured with one or more first operation instructions. The first operation instructions configured for each user-defined instruction can be the same or different.

[0072] Exemplarily, the user may set a custom instruction and its first corresponding relationship in an application in the vehicle-mounted device.

[0073] In one example, the user sets the user-defined instruction to "Hello White Rabbit", and the corresponding first instruction operation may be "Turn on the air conditioner".

[0074] In some embodiments, in S180, the vehicle-mounted device may execute the first operation instruction to implement the device function corresponding to the first operation instruction.

[0075] In one example, the first operation instruction may be “turn on the air conditioner”, and the vehicle-mounted device executing the first operation instruction may be the vehicle-mounted device turning on the air conditioner.

[0076] In one example, the user can set the user-defined instruction to "Hello White Rabbit", and its corresponding first instruction operation can be "Turn on the air conditioner". In addition, after the vehicle-mounted device recognizes the user's first audio, the first audio and the user-defined instruction can be sent to the server. After receiving the first audio and the user-defined instruction, the server can determine the credibility threshold corresponding to the user instruction through the audio recognition model. For example, the credibility threshold corresponding to "Hello White Rabbit" is 0.6. In addition, the first audio and the user-defined instruction are matched by the audio recognition model, and the probability that the first audio is recognized as "Hello White Rabbit" is 0.7. The server can return the credibility threshold and the matching degree to the vehicle-mounted device. After receiving the credibility threshold and the matching degree, the vehicle-mounted device can determine whether the matching degree is greater than the credibility threshold. After determining that the matching degree is 0.7 and is greater than the credibility threshold 0.6, the vehicle-mounted device can control the air conditioner to turn on.

[0077] In the embodiment of the present application, after the vehicle-mounted device receives the first voice command, the first voice command and the user-defined command can be transmitted to the server. After receiving the first voice command and the user-defined command, the server can use the internally deployed audio recognition model to determine the matching degree between the first voice command and the user-defined command transmission, and feed back the matching degree to the vehicle-mounted device, and the vehicle-mounted device makes a decision based on the credibility threshold and matching degree corresponding to the user-defined command, that is, when it is determined that the matching degree is greater than or equal to the credibility threshold, the first operation instruction corresponding to the user-defined instruction is determined and executed according to the first corresponding relationship. It can be understood that in the embodiment of the present application, through the user-defined instruction set by the user on the vehicle-mounted device end, each time after receiving the first voice instruction, the corresponding vehicle-mounted device uploads, receives and arbitrates, so that different vehicle-mounted devices can execute corresponding operation instructions according to the user-defined instructions set by different users, and different vehicle-mounted devices are affected by each other, so as to realize user-level custom instruction control.

[0078] In order to ensure the versatility and universality of device control, as another implementation method of the present application, the present application also provides another implementation method of voice command processing, please refer to the following embodiments for details.

[0079] Figure 2 FIG. 2 is a flow chart showing a method for processing voice commands according to another embodiment of the present invention. Figure 2 As shown, the voice command processing method includes the following steps:

[0080] S201: The vehicle-mounted device receives a first voice command.

[0081] S202: The vehicle-mounted device sends the first voice command and the user-defined command stored in the vehicle-mounted device to the server.

[0082] S203: The server receives a first voice command and a user-defined command sent by the vehicle-mounted device.

[0083] S204: The server calculates a matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain a matching degree between the first voice instruction and the user-defined instruction.

[0084] S205. The server converts the first voice instruction into text through an audio recognition model to obtain text information corresponding to the first voice instruction.

[0085] S206: The server sends the matching degree and text information to the vehicle-mounted device.

[0086] S207: The in-vehicle device receives the matching degree between the first voice command and the user-defined command and the text information corresponding to the first voice command.

[0087] S208. The vehicle-mounted device determines whether the matching degree is greater than or equal to the credibility threshold corresponding to the custom instruction. If so, execute step S209; if not, execute step S211.

[0088] S209: The vehicle-mounted device determines a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, where the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions.

[0089] S210: The vehicle-mounted device executes a first operation instruction.

[0090] S212: The vehicle-mounted device determines a second operation instruction corresponding to the text information according to the text information.

[0091] S212: The vehicle-mounted device executes the second operation instruction.

[0092] Exemplarily, steps S201-S204 are consistent with steps S110-S140, and steps S209-S210 are consistent with steps S170-S180, which are not elaborated herein.

[0093] In some embodiments, in S205, the server may perform text recognition on the first voice instruction through an audio recognition model to obtain text information corresponding to the first voice instruction.

[0094] In one example, after the model receives the first voice instruction, the first voice instruction can be preprocessed. Among them, the preprocessing of the first voice instruction can include operations such as sampling, quantization and framing. For example, the audio signal of the first voice instruction is discretized at a certain sampling rate (for example, 16kHz), and the length of each frame of audio can be set to about 20-20 milliseconds, and there is a partial overlap between adjacent frames (such as 10-15 milliseconds of overlap). At the same time, the audio may be normalized so that the amplitude range of the audio signal meets the input requirements of the model. In addition, after preprocessing the first voice instruction, acoustic features can be extracted from each frame of audio. Among them, the acoustic features can be extracted by Mel frequency cepstrum coefficients. The extraction process is to first perform a fast Fourier transform (FFT) on the audio frame, convert the time domain signal into a frequency domain signal, and then filter the frequency domain signal through a Mel filter bank, and finally take the logarithm of the filtered result and perform a discrete cosine transform (DCT) to obtain the MFCC feature vector. These feature vectors are used as input to the speech recognition model. Furthermore, the extracted acoustic features may be used for a voice command recognition model, and the model outputs a text sequence corresponding to the first voice command, that is, outputs text information corresponding to the first voice command.

[0095] In some embodiments, in S206 , the server sends the credibility threshold, matching degree, and text information determined by the audio recognition model to the in-vehicle device.

[0096] In some optional embodiments, Figure 3 FIG. 1 is a flow chart showing a text message sending process provided by an embodiment of the present application. Figure 3 As shown, the voice command processing method includes the following steps:

[0097] S2061. Determine the credibility score corresponding to the text information through an audio recognition model.

[0098] In one example, an audio recognition model can be designed with an output layer to predict the text character or word corresponding to the speech feature corresponding to each time step. A softmax function or other probability distribution function can be deployed in the output layer of the audio recognition model to output the probability of each possible character or word. Furthermore, the confidence score corresponding to the text information can be obtained by calculating the product of the probabilities of all time steps or words, or calculating the sum of logarithmic probabilities.

[0099] S2062: When it is determined that the credibility score is greater than or equal to a preset credibility threshold, the text information corresponding to the first voice command is sent to the vehicle-mounted device.

[0100] For example, when the server determines that the credibility score is greater than a preset credibility threshold, the server can send the text message corresponding to the first voice command to the vehicle-mounted device. The preset credibility threshold can be pre-set by a technician according to different requirements.

[0101] It can be understood that the credibility score can be used to characterize the degree of confidence of the audio recognition model in the prediction results. Therefore, by determining the credibility score of the text information corresponding to the first voice command, and only sending the text information to the vehicle-mounted device when it is determined that the credibility score meets the preset credibility threshold, the text information can be screened by the credibility score, which can improve the accuracy of the text recognition results, thereby improving the accuracy and reliability of audio recognition, and ensuring that the vehicle-mounted device can receive accurate text information.

[0102] In some embodiments, in S207, after determining the credibility threshold, the matching degree and the text information, the vehicle-mounted device may execute step S208, i.e., determine whether the matching degree is greater than the credibility threshold. If it is determined that the matching degree is greater than or equal to the credibility threshold, steps S209 and S210 are executed, i.e., according to the first corresponding relationship, the first operation instruction corresponding to the user-defined instruction is determined and executed. If it is determined that the matching degree is less than the credibility threshold, steps S211 and S212 may be executed. That is, the vehicle-mounted device determines the second operation instruction corresponding to the text information according to the text information; the vehicle-mounted device executes the second operation instruction.

[0103] Exemplarily, the vehicle-mounted device may determine the second operation instruction according to the instruction word in the text message. In one example, the text message may be "Hello, please help me turn on the air conditioner", and it may be determined that the second operation instruction may be to turn on the air conditioner.

[0104] In an embodiment of the present application, when the degree of match is less than a credibility threshold, the vehicle-mounted device can execute the second operation instruction corresponding to the text information. This ensures that regardless of whether the user sets a custom instruction, the vehicle-mounted device can implement the function of the vehicle-mounted device corresponding to the user instruction, thereby improving the universality of voice control.

[0105] Furthermore, in order to determine the credibility threshold corresponding to the user-defined instruction, as another implementation method of the present application, the present application also provides another implementation method of voice instruction processing, please refer to the following embodiments for details.

[0106] Figure 4FIG. 2 is a flow chart showing a method for processing voice commands according to another embodiment of the present invention. Figure 4 As shown, the voice command processing method includes the following steps:

[0107] S410: The vehicle-mounted device sends the user-defined instruction to the server.

[0108] S420: The server receives a user-defined instruction.

[0109] S430: The server determines a credibility threshold corresponding to the user-defined instruction through an audio recognition model.

[0110] S440: The server sends the credibility threshold to the vehicle-mounted device.

[0111] S450: The vehicle-mounted device receives a credibility threshold corresponding to the user-defined instruction.

[0112] In some embodiments, in S410, the vehicle-mounted device may send the user-defined instruction to the server. And in S420, the server receives the user-defined instruction sent by the vehicle-mounted device.

[0113] In some embodiments, in S430, after receiving the user-defined instruction, the server may determine the credibility threshold corresponding to the user-defined instruction through an audio recognition model deployed in the server.

[0114] Exemplarily, the server may use a user-defined instruction voice instruction recognition model to analyze the user-defined instruction through an audio recognition model, thereby determining a credibility threshold corresponding to the user-defined instruction.

[0115] In some optional embodiments, Figure 5 FIG. 2 shows a flow chart of determining a credibility threshold provided by an embodiment of the present application. Figure 5 As shown, the voice command processing method includes the following steps:

[0116] S431, determining the pinyin sequence corresponding to the user-defined instruction.

[0117] Exemplarily, the pinyin sequence corresponding to the user-defined instruction can be obtained through an audio recognition model.

[0118] In one example, the audio recognition model can split the user-defined instructions, that is, perform word segmentation on the user-defined instructions to obtain user-defined instruction segments. Among them, the audio recognition model can divide the user-defined instructions according to factors such as voice rhythm and pauses, so as to better match the voice segments when actually speaking. In addition, the user-defined instruction segments can include words or phrases. The word segmentation of user-defined instructions can be achieved through rules or algorithms such as maximum matching method and hidden Markov model (HMM) word segmentation.

[0119] Furthermore, each user-defined instruction segment can be converted to pinyin. For example, the pinyin conversion can be achieved through a pre-built Chinese character-pinyin mapping table. In addition, when there are polyphones in the user-defined instruction segment, the correct pronunciation of the polyphone can be determined by combining the context.

[0120] Furthermore, the pinyin corresponding to multiple user-defined instruction segments may be integrated to obtain a pinyin sequence corresponding to the user-defined instruction.

[0121] S432. Analyze the pinyin sequence through the audio recognition model to determine the credibility threshold corresponding to the pinyin sequence.

[0122] Exemplarily, the determined pinyin sequence voice instruction recognition model is used to analyze and identify the pinyin sequence through an audio recognition model, thereby determining a credibility threshold corresponding to the pinyin sequence.

[0123] In one example, the audio recognition model can be trained by the following steps: obtain an audio sample set, wherein the audio sample set includes multiple pinyin samples and their corresponding labels. Input the audio sample into a preset model, wherein the preset model can be a classification model, such as a decision tree. Perform feature extraction on each pinyin sample through the preset model to obtain features such as pinyin length and pinyin combination frequency. Input the extracted features into the preset model, classify the pinyin sample features, and obtain predicted classification results. Calculate the loss function value using the predicted classification results and labels. When it is determined that the loss function value does not meet the preset stop condition, adjust the parameters of the preset model, and return to the step of inputting the audio sample into the preset model until the loss function value meets the preset stop condition.

[0124] Furthermore, the preset model can output a probability or confidence level to determine the credibility of the pinyin sequence. By adjusting the threshold, the accuracy and recall rate of the model can be balanced. Among them, the optimal threshold can be found by methods such as cross-validation and grid search.

[0125] In the embodiments of the present application, it can be understood that Pinyin has clear syllables and pronunciation rules. Therefore, by converting user-defined instructions into Pinyin sequences, the model can better extract and represent the features of user-defined instructions, and thus determine a more accurate credibility threshold.

[0126] Further, in some embodiments, in S440, the server may send the credibility threshold to the vehicle-mounted device. And in S450, the vehicle-mounted device receives and stores the credibility threshold corresponding to the user-defined instruction.

[0127] In an embodiment of the present application, the credibility threshold corresponding to the user-defined instruction can be determined by the audio recognition model in the server, and the vehicle-mounted device can receive the credibility threshold sent by the server and store it in the vehicle-mounted device, so that the vehicle-mounted device can directly obtain the corresponding credibility threshold according to the user-defined instruction, thereby improving the judgment efficiency of user instructions.

[0128] Exemplarily, voice command processing is described in conjunction with the following examples.

[0129] In one example, the user can set the user-defined instruction to "Hello White Rabbit", and the corresponding first instruction operation can be "Turn on the air conditioner". In addition, after the in-vehicle device recognizes the user's first audio, the first audio and the user-defined instruction can be sent to the server. After receiving the first audio and the user-defined instruction, the server can determine the credibility threshold corresponding to the user instruction through the audio recognition model. For example, the credibility threshold corresponding to "Hello White Rabbit" is 0.6. In addition, the audio recognition model is used to calculate the matching degree of the first audio and the user-defined instruction, and the probability that the first audio is recognized as "Hello White Rabbit" is 0.12. In addition, the audio recognition model performs text recognition on the first audio and the result can be "Turn on the air conditioner", and its corresponding credibility score is 0.75.

[0130] The server can return the credibility threshold, matching degree, text recognition results and their corresponding credibility scores to the vehicle-mounted device. After receiving the credibility threshold and matching degree, the vehicle-mounted device can determine whether the matching degree is greater than the credibility threshold. After determining that the matching degree is 0.12 and is greater than the credibility threshold 0.6, the operation instruction to turn on the air conditioner can be executed.

[0131] In another example, the audio recognition model calculates the matching degree of the first audio and the user-defined instruction, and obtains a probability of 0.7 that the first audio is recognized as "Hello Big White Rabbit". In addition, the text recognition result of the audio recognition model on the first audio can be "Hello Little White Rabbit", and its corresponding credibility score is 0.75. After receiving the credibility threshold and the matching degree, the vehicle-mounted device can determine whether the matching degree is greater than the credibility threshold. After determining that the matching degree is 0.7 and is greater than the credibility threshold of 0.6, the operation instruction corresponding to "Hello Big White Rabbit" can be executed, that is, turning on the air conditioner.

[0132] Based on the voice instruction processing method provided in the above embodiment, the present application also provides a specific implementation of a voice instruction processing device. Please refer to the following embodiment.

[0133] See first Figure 6 The voice instruction processing device 600 provided in the embodiment of the present application includes the following units:

[0134] A first receiving module 601, used to receive a first voice instruction;

[0135] A sending module 602, configured to send the first voice command and the user-defined command stored in the vehicle-mounted device to a server in which an audio recognition model is deployed;

[0136] A second receiving module 603 is used to receive a matching degree between the first voice instruction and the user-defined instruction, where the matching degree is generated based on the audio recognition model;

[0137] A determination module 604 is configured to determine, when it is determined that the matching degree is greater than or equal to a credibility threshold corresponding to the user-defined instruction, a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, wherein the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions;

[0138] The execution module 605 is used to execute the first operation instruction.

[0139] As an implementation method of the present application, in order to improve the versatility and universality of device control, the second receiving module 603 is also used to: receive text information corresponding to the first voice instruction, the text information is generated based on the audio recognition model; the determination module 604 is also used to determine the second operation instruction corresponding to the text information based on the text information when it is determined that the matching degree is less than the credibility threshold corresponding to the custom instruction; the execution module 605 is also used to execute the second operation instruction.

[0140] As an implementation method of the present application, the sending module 602 is also used to send the user-defined instruction to the server, so that the server determines the credibility threshold corresponding to the user-defined instruction based on the audio recognition model; the second receiving module 603 is also used to receive the credibility threshold corresponding to the user-defined instruction.

[0141] See also Figure 7 The voice command processing device 700 provided in the embodiment of the present application includes the following units:

[0142] The receiving module 701 is used to receive a first voice command sent by the vehicle-mounted device;

[0143] A first determination module 702, configured to calculate a matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain a matching degree between the first voice instruction and the user-defined instruction;

[0144] The sending module 703 is used to send the matching degree to the vehicle-mounted device, so that when the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, it determines the first operation instruction corresponding to the user-defined instruction according to the first corresponding relationship and executes the first operation instruction.

[0145] As an implementation method of the present application, the receiving module 701 is also used to receive the user-defined instruction; the first determination module 702 is also used to determine the credibility threshold corresponding to the user-defined instruction through the audio recognition model; the sending module 703 is also used to send the credibility threshold to the vehicle-mounted device.

[0146] As an implementation method of the present application, the first determination module 702 determines the credibility threshold corresponding to the user-defined instruction through an audio recognition model in the following manner: determining the pinyin sequence corresponding to the user-defined instruction; analyzing the pinyin sequence through an audio recognition model to determine the credibility threshold corresponding to the pinyin sequence.

[0147] As an implementation method of the present application, after receiving the first voice command sent by the vehicle and the user-defined command, the device also includes a third determination module: used to convert the first voice command into text through an audio recognition model to obtain text information corresponding to the first voice command; the sending module 704 is also used to send the text information corresponding to the first voice command to the vehicle-mounted device, so that the vehicle-mounted device determines the second operation instruction corresponding to the text information based on the text information when it is determined that the matching degree is less than the credibility threshold; and executes the second operation instruction.

[0148] As an implementation method of the present application, the sending module 703 sends the text information corresponding to the first voice command to the vehicle-mounted device in the following manner: determine the credibility score corresponding to the text information through the audio recognition model; when it is determined that the credibility score is greater than or equal to the preset credibility threshold, send the text information corresponding to the first voice command to the vehicle-mounted device.

[0149] In the embodiment of the present application, after the vehicle-mounted device receives the first voice command, the first voice command and the user-defined command pre-stored in the vehicle-mounted device can be transmitted to the server. After receiving the first voice command and the user-defined command, the server can use the internally deployed audio recognition model to determine the matching degree between the first voice command and the user-defined command transmission, and feed back the matching degree to the vehicle-mounted device, and the vehicle-mounted device makes a decision based on the credibility threshold and matching degree corresponding to the user-defined command, that is, when it is determined that the matching degree is greater than or equal to the credibility threshold, the first operation instruction corresponding to the user-defined instruction is determined and executed according to the first corresponding relationship. It can be understood that in the embodiment of the present application, through the user-defined instruction set by the user on the vehicle-mounted device end, each time after receiving the first voice instruction, the corresponding vehicle-mounted device uploads, receives and arbitrates, so that different vehicle-mounted devices can execute corresponding operation instructions according to the user-defined instructions set by different users, and different vehicle-mounted devices have different effects on each other, thereby realizing user-level custom instruction control.

[0150] Figure 8 A schematic diagram of the hardware structure of the voice command processing provided in an embodiment of the present application is shown.

[0151] The voice instruction processing device may include a processor 801 and a memory 802 storing computer program instructions.

[0152] Specifically, the processor 801 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0153] The memory 802 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 802 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 802 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 802 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 802 is a non-volatile solid-state memory.

[0154] The memory may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical or other physical / tangible memory storage devices. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0155] The processor 801 implements any one of the voice instruction processing methods in the above embodiments by reading and executing the computer program instructions stored in the memory 802 .

[0156] In one example, the voice instruction processing device may further include a communication interface 803 and a bus 810. Figure 8 As shown, the processor 801, the memory 802, and the communication interface 803 are connected via a bus 810 and communicate with each other.

[0157] The communication interface 803 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0158] Bus 810 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 810 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0159] The voice instruction processing device can execute the voice instruction processing method in the embodiment of the present application based on the first voice instruction and the user-defined instruction, thereby realizing the combination of Figure 1 , Figure 5 and Figure 6 The described voice command processing method and device.

[0160] In addition, in combination with the voice instruction processing method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any one of the voice instruction processing methods in the above embodiment is implemented.

[0161] The present embodiment further provides a computer program product, including a computer program, which implements any one of the voice instruction processing methods in the above embodiments when the computer program is executed by a processor.

[0162] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0163] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0164] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0165] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0166] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A method for processing voice commands, characterized in that: Applied to an in-vehicle device, the method comprises: receiving a first voice command; Sending the first voice command and the user-defined command stored in the vehicle-mounted device to a server, wherein an audio recognition model is deployed in the server; Receiving a matching degree between the first voice instruction and the user-defined instruction, wherein the matching degree is generated based on the audio recognition model; In the case where it is determined that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, determining a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, wherein the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions; Execute the first operation instruction.

2. The method according to claim 1, characterized in that: After sending the first voice instruction and the user-defined instruction to the server, the method further includes: receiving text information corresponding to the first voice instruction, wherein the text information is generated based on the audio recognition model; In the case where it is determined that the matching degree is less than the credibility threshold corresponding to the user-defined instruction, determining a second operation instruction corresponding to the text information according to the text information; Execute the second operation instruction.

3. The method according to claim 1, characterized in that: Before receiving the first voice instruction, the method further includes: Sending the user-defined instruction to a server, so that the server determines a credibility threshold corresponding to the user-defined instruction according to the audio recognition model; A credibility threshold corresponding to the user-defined instruction is received.

4. A method for processing voice commands, characterized in that: Applied to a server, the method comprises: Receiving a first voice command and a user-defined command sent by the vehicle-mounted device; Calculating the matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain the matching degree between the first voice instruction and the user-defined instruction; The matching degree is sent to the vehicle-mounted device, so that when the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, the vehicle-mounted device determines the first operation instruction corresponding to the user-defined instruction according to the first corresponding relationship and executes the first operation instruction.

5. The method according to claim 4, characterized in that The method further comprises: Receiving the user-defined instruction; Determining a credibility threshold corresponding to the user-defined instruction through an audio recognition model; The credibility threshold is sent to the vehicle-mounted device.

6. The method according to claim 5, characterized in that The step of determining the credibility threshold corresponding to the user-defined instruction by using the audio recognition model includes: Determine the pinyin sequence corresponding to the user-defined instruction; The pinyin sequence is analyzed by the audio recognition model to determine a credibility threshold corresponding to the pinyin sequence.

7. The method according to claim 4, characterized in that After receiving the first voice command sent by the vehicle and the user-defined command, the method further includes: Convert the first voice command into text using an audio recognition model to obtain text information corresponding to the first voice command; The text information corresponding to the first voice instruction is sent to the vehicle-mounted device, so that when the vehicle-mounted device determines that the matching degree is less than the credibility threshold, the vehicle-mounted device determines a second operation instruction corresponding to the text information based on the text information and executes the second operation instruction.

8. The method according to claim 7, characterized in that The sending of the text information corresponding to the first voice command to the vehicle-mounted device includes: Determining the credibility score corresponding to the text information through the audio recognition model; When it is determined that the credibility score is greater than or equal to a preset credibility threshold, the text information corresponding to the first voice instruction is sent to the vehicle-mounted device.

9. A voice command processing device, characterized in that: Applied to vehicle-mounted equipment, the device comprises: A first receiving module, used to receive a first voice command; A sending module, used for sending the first voice instruction and the user-defined instruction stored in the vehicle-mounted device to a server, in which an audio recognition model is deployed; A second receiving module, used to receive a matching degree between the first voice instruction and the user-defined instruction, wherein the matching degree is generated based on the audio recognition model; a determination module, configured to determine, when it is determined that the matching degree is greater than or equal to a credibility threshold corresponding to the user-defined instruction, a first operation instruction corresponding to the user-defined instruction according to a first corresponding relationship, wherein the first corresponding relationship represents a corresponding relationship between the user-defined instruction and one or more operation instructions; An execution module is used to execute the first operation instruction.

10. A voice command processing device, characterized in that: Applied to a server, the device comprises: A receiving module, used for receiving a first voice command sent by the vehicle-mounted device; A first determination module, configured to calculate a matching degree between the first voice instruction and the user-defined instruction through an audio recognition model to obtain a matching degree between the first voice instruction and the user-defined instruction; A sending module is used to send the matching degree to the vehicle-mounted device, so that when the vehicle-mounted device determines that the matching degree is greater than or equal to the credibility threshold corresponding to the user-defined instruction, the vehicle-mounted device determines the first operation instruction corresponding to the user-defined instruction according to the first corresponding relationship and executes the first operation instruction.

11. A voice command processing device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, it implements the voice instruction processing method as described in any one of claims 1-3; or it implements the voice instruction processing method as described in any one of claims 4-8.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by the processor, the voice instruction processing method as described in any one of claims 1-3 is implemented; or the voice instruction processing method as described in any one of claims 4-8 is implemented.

13. A computer program product, characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the voice instruction processing method as described in any one of claims 1-3; or implements the voice instruction processing method as described in any one of claims 4-8.

Images