Speech recognition method, device, server and electronic equipment

Abstract

The present application discloses a speech recognition method, device, server, and electronic device, which relate to the field of artificial intelligence technology. The method solves the problem in the prior art that when a television receives unclear or ambiguous speech, the television cannot determine the user's intention based on the speech, and then the television defaults the user's intention to the default intention, resulting in a low recognition rate of the actual user's intention. The method includes: receiving query information sent by an electronic device; inputting the speech data into a dialogFlow model, and when it is determined that the user's intention is the default intention, inputting the speech data into a pre-configured text convolutional network model to determine the intent score of each intent except the default intent; and when it is determined that there is an intent with an intent score greater than a first threshold, sending first display information containing the intent with an intent score greater than the first threshold to the electronic device.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and more specifically, to a speech recognition method, device, server, and electronic device. Background Art

[0002] With the development of voice recognition technology, intelligent voice interaction technology has gradually become a standard feature of terminal devices (such as mobile phones, tablets, or smart home products such as smart appliances). In intelligent voice interaction scenarios, users can control smart home appliances through voice. Taking the terminal device as a TV as an example, users can control the TV through voice to perform a series of TV control operations such as watching videos, listening to music, or checking the weather. However, under normal circumstances, when the TV receives unclear or ambiguous voice, the TV cannot determine the user's intention based on the voice. The TV will then default to the user's intention, resulting in a low recognition rate of the actual user's intention. Summary of the Invention

[0003] The present application provides a speech recognition method, device, server and electronic device, which solves the problem in the prior art that when a television receives unclear or ambiguous speech, the television cannot determine the user's intention based on the speech, and then the television defaults to the user's intention, resulting in a low recognition rate of the actual user intention.

[0004] To achieve the above objectives, this application adopts the following technical solutions:

[0005] In a first aspect, the present application provides a speech recognition method, comprising: receiving query information sent by an electronic device; the query information includes voice data; when the voice data is input into a dialogFlow model and it is determined that the user intention is a default intention, the voice data is input into a pre-configured text convolutional network model to determine the intention score of each intention except the default intention; when it is determined that there is an intention with an intention score greater than a first threshold, first display information containing the intention with an intention score greater than the first threshold is sent to the electronic device, the first display information being used to instruct the electronic device to execute the intention with an intention score greater than the first threshold and search for voice data.

[0006] In some feasible examples, the speech recognition method provided by the present application also includes: obtaining historical query data corresponding to at least one default intent; the historical query data includes at least one query information, and one query information in the at least one query information corresponds to a default intent; inputting the speech data corresponding to each query information in the at least one query information into the text convolutional network model, determining the maximum intent score corresponding to each default intent, and the intent corresponding to the maximum intent score; determining the first threshold and the second threshold based on the maximum intent score corresponding to each default intent, and the intent corresponding to the maximum intent score.

[0007] In some feasible examples, the speech recognition method provided in the present application also includes: when the intention scores are all less than or equal to the first threshold, determining the intention whose intention score is greater than or equal to the second threshold; sending a second display information containing the intention whose intention score is greater than or equal to the second threshold to the electronic device, and the second display information is used to instruct the electronic device to display the intention whose intention score is greater than or equal to the second threshold.

[0008] In some feasible examples, the speech recognition method provided in the present application also includes: when it is determined that the intention scores are all less than a second threshold, sending a third display information containing the default intention to the electronic device, and the third display information is used to instruct the electronic device to execute the default intention and search for voice data.

[0009] In some feasible examples, the query information also includes an international language code; inputting the speech data into a preconfigured text convolutional network model to determine the intent score for each intent except the default intent, including: when the language corresponding to the international language code is a target language, performing character classification preprocessing on the speech data to determine first classification data; wherein the target language includes one or more of Chinese, Thai, and Japanese. Inputting the first classification data into the preconfigured text convolutional network model to determine the intent score for each intent except the default intent.

[0010] In some feasible examples, the query information also includes an international language code; the speech data is input into a preconfigured text convolutional network model to determine the intent score for each intent except the default intent, including: when the language corresponding to the international language code is not a target language, performing word classification preprocessing on the speech data to determine second classification data; wherein the target language includes one or more of Chinese, Thai, and Japanese. The second classification data is input into the preconfigured text convolutional network model to determine the intent score for each intent except the default intent.

[0011] In some feasible examples, the training process of the text convolutional network model is as follows: obtain training sentences and the marking results of the training sentences; wherein the training sentences include voice data and intent; input the training sentences into the neural network model; based on the target loss function, determine whether the prediction results of the training sentences output by the neural network model match the marking results; if the prediction results do not match the marking results, iterate the configuration parameters of the neural network model in a loop until the neural network model converges to obtain the text convolutional network model.

[0012] In some feasible examples, the speech recognition method provided by the present application also includes: inputting the speech data into a preconfigured text convolutional network model to determine the confidence of each intent except the default intent; and determining the intent score corresponding to each intent except the default intent based on the confidence.

[0013] In some feasible examples, the speech recognition method provided by the present application also includes: obtaining the usage frequency of the intent category to which each intent belongs except the default intent; determining the intent coefficient corresponding to each intent category based on the usage frequency; inputting the speech data into a pre-configured text convolutional network model to determine the confidence of each intent except the default intent; determining the intent score corresponding to each intent except the default intent based on the confidence and the intent coefficient corresponding to each intent category.

[0014] In a second aspect, the present application provides a speech recognition method, which is applied to an electronic device, including: sending query information to a server; the query information includes voice data; receiving first display information sent by the server containing an intention with an intention score greater than a first threshold; according to the first display information, executing the intention with an intention score greater than the first threshold, and searching for voice data; displaying the execution of the intention with an intention score greater than the first threshold, and searching for search results of the voice data.

[0015] In some feasible examples, the speech recognition method provided in the present application also includes: receiving second display information sent by the server containing an intent whose intent score is greater than or equal to a second threshold; and displaying the intent whose intent score is greater than or equal to the second threshold according to the second display information.

[0016] In some feasible examples, displaying intents with intent scores greater than or equal to a second threshold based on the second display information includes: sorting the intent scores in descending order based on the second display information, and determining the display order of each intent with an intent score greater than or equal to the second threshold; and displaying intents with intent scores greater than or equal to the second threshold based on the display order.

[0017] In some feasible examples, the speech recognition method provided by the present application also includes: receiving third display information containing a default intent sent by a server; executing the default intent according to the third display information, and searching for voice data.

[0018] In a third aspect, the present application provides a speech recognition device, comprising: a transceiver unit for receiving query information sent by an electronic device; the query information includes voice data; a processing unit for inputting the voice data received by the receiving unit into a dialogFlow model, and determining that the user intention is a default intention, and then inputting the voice data into a pre-configured text convolutional network model to determine the intention score of each intention except the default intention; the processing unit is also used to control the transceiver unit to send a first display information containing the intention score greater than the first threshold to the electronic device when determining that there is an intention with an intention score greater than a first threshold, the first display information being used to instruct the electronic device to execute the intention with an intention score greater than the first threshold and search for voice data.

[0019] In some feasible examples, the transceiver unit is further used to obtain historical query data corresponding to at least one default intent; the historical query data includes at least one query information, and one query information in the at least one query information corresponds to a default intent; the processing unit is further used to input the voice data corresponding to each query information in the at least one query information obtained by the transceiver unit into the text convolutional network model, and determine the maximum intent score corresponding to each default intent, and the intent corresponding to the maximum intent score;

[0020] The processing unit is further configured to determine a first threshold and a second threshold based on a maximum intent score corresponding to each default intent and an intent corresponding to the maximum intent score.

[0021] In some feasible examples, the processing unit is further used to determine an intention whose intention score is greater than or equal to a second threshold when all intention scores are less than or equal to the first threshold; the processing unit is further used to control the transceiver unit to send a second display message containing an intention whose intention score is greater than or equal to the second threshold to the electronic device, and the second display information is used to instruct the electronic device to display the intention whose intention score is greater than or equal to the second threshold.

[0022] In some feasible examples, the processing unit is also used to control the transceiver unit to send a third display message containing the default intent to the electronic device when determining that the intent scores are all less than a second threshold. The third display information is used to instruct the electronic device to execute the default intent and search for voice data.

[0023] In some feasible examples, the query information also includes an international language code; the processing unit is specifically used to perform character classification preprocessing on the voice data when the language corresponding to the international language code received by the transceiver unit is the target language, and determine first classification data; wherein the target language includes one or more of Chinese, Thai and Japanese; the processing unit is specifically used to input the first classification data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent.

[0024] In some feasible examples, the query information also includes an international language code; the processing unit is specifically used to perform word classification preprocessing on the voice data to determine second classification data when the language corresponding to the international language code received by the transceiver unit is not the target language; wherein the target language includes one or more of Chinese, Thai and Japanese; the processing unit is specifically used to input the second classification data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent.

[0025] In some feasible examples, the training process of the text convolutional network model is as follows: the transceiver unit is also used to obtain training sentences and the marking results of the training sentences; wherein the training sentences include voice data and intent; the processing unit is also used to input the training sentences obtained by the transceiver unit into the neural network model; the processing unit is also used to determine whether the prediction results of the training sentences output by the neural network model match the marking results based on the target loss function; the processing unit is also used to cyclically iterate the configuration parameters of the neural network model when the prediction results do not match the marking results until the neural network model converges to obtain the text convolutional network model.

[0026] In some feasible examples, the processing unit is further used to input the voice data received by the transceiver unit into a preconfigured text convolutional network model to determine the confidence of each intent except the default intent; the processing unit is further used to determine the intent score corresponding to each intent except the default intent based on the confidence.

[0027] In some feasible examples, the transceiver unit is further used to obtain the usage frequency of the intent category to which each intent belongs except the default intent; the processing unit is further used to determine the intent coefficient corresponding to each intent category based on the usage frequency obtained by the transceiver unit; the processing unit is further used to input the voice data into a pre-configured text convolutional network model to determine the confidence of each intent except the default intent; the processing unit is further used to determine the intent score corresponding to each intent except the default intent based on the confidence and the intent coefficient corresponding to each intent category.

[0028] In a fourth aspect, the present application provides a speech recognition device, comprising: a transceiver unit for sending query information to a server; the query information includes voice data; the transceiver unit is also used to receive first display information sent by the server containing an intention with an intention score greater than a first threshold; the processing unit is also used to execute the intention with an intention score greater than the first threshold and search for voice data based on the first display information received by the transceiver unit; and the display unit is used to display the search results of the processing unit executing the intention with an intention score greater than the first threshold and searching for voice data.

[0029] In some feasible examples, the transceiver unit is further used to receive second display information sent by the server containing an intention whose intention score is greater than or equal to a second threshold; the processing unit is further used to control the display unit to display the intention whose intention score is greater than or equal to the second threshold based on the second display information received by the transceiver unit.

[0030] In some feasible examples, the processing unit is specifically used to sort the intention scores in descending order according to the second display information received by the transceiver unit, and determine the display order of each intention whose intention score is greater than or equal to the second threshold; the processing unit is specifically used to control the display unit to display the intentions whose intention scores are greater than or equal to the second threshold according to the display order.

[0031] In some feasible examples, the transceiver unit is further used to receive a third display message containing a default intent sent by the server; the processing unit is further used to execute the default intent and search for voice data based on the third display message received by the transceiver unit; the display unit is further used to display the search results of the processing unit executing the default intent and searching for voice data.

[0032] In a fifth aspect, the present application provides a speech recognition system, characterized in that it includes a server and an electronic device, the server executes any speech recognition method provided in the first aspect, and the electronic device executes any speech recognition method provided in the second aspect.

[0033] In a sixth aspect, the present application provides a computer-readable storage medium, characterized in that it includes instructions that, when run on a computer, enable the computer to execute any one of the speech recognition methods provided in the first aspect.

[0034] In the seventh aspect, the present application provides a server, characterized in that it includes: a communication interface, a processor, a memory, and a bus; the memory is used to store computer-executable instructions, and the processor is connected to the memory through the bus; when the server is running, the processor executes the computer-executable instructions stored in the memory, so that the server executes the speech recognition method as described in any one of the items provided in the first aspect.

[0035] In an eighth aspect, the present application provides a computer program product, which, when executed on a computer, enables the computer to execute the speech recognition method as described in the design method of the first aspect.

[0036] In the ninth aspect, the present application provides a computer-readable storage medium, characterized in that it includes instructions that, when run on a computer, enable the computer to execute the speech recognition method as described in any one of the second aspects.

[0037] In the tenth aspect, the present application provides an electronic device, characterized in that it includes: a communication interface, a processor, a memory, and a bus; the memory is used to store computer execution instructions, and the processor is connected to the memory through the bus; when the electronic device is running, the processor executes the computer execution instructions stored in the memory, so that the electronic device performs the speech recognition method as described in any one of the items provided in the second aspect.

[0038] In an eleventh aspect, the present application provides a computer program product, which, when executed on a computer, enables the computer to execute the speech recognition method as described in the design method of the second aspect.

[0039] It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a first computer-readable storage medium. The first computer-readable storage medium may be packaged together with the server, or may be packaged separately with the electronic device or the server processor, and this application does not limit this.

[0040] The descriptions of the third, sixth, seventh and eighth aspects of this application can refer to the detailed description of the first aspect; and the beneficial effects of the descriptions of the third, sixth, seventh and eighth aspects can refer to the analysis of the beneficial effects of the first aspect, which will not be repeated here.

[0041] The descriptions of the fourth, ninth, tenth and eleventh aspects of this application can refer to the detailed description of the second aspect; and the beneficial effects of the descriptions of the fourth, ninth, tenth and eleventh aspects can refer to the beneficial effect analysis of the second aspect, which will not be repeated here.

[0042] In this application, the names of the aforementioned servers or electronic devices do not limit the devices or functional modules themselves. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they fall within the scope of this application and its equivalents.

[0043] These and other aspects of the present application will become more readily apparent from the following description.

[0044] The technical solution provided by this application has the following advantages compared with the existing technology:

[0045] After the voice data sent by the electronic device is input into the dialogFlow model, if the dialogFlow model cannot recognize the voice data, the user intention of the dialogFlow model is the default intention, resulting in a low recognition rate of the actual user intention. To this end, the speech recognition method provided by the present application, when the dialogFlow model determines that the user intention is the default intention, by re-analyzing the voice data sent by the electronic device, it is possible to determine the intention score of each intention except the default intention, and further determine the user intention based on the relationship between the intention score and the first threshold. For example, when it is determined that there is an intention with an intention score greater than the first threshold, the first display information containing the intention with an intention score greater than the first threshold is sent to the electronic device. In this way, the recognition rate of the actual user intention can be improved, solving the problem in the prior art that when a television receives an unclear or ambiguous voice, the television cannot determine the user's user intention based on the voice, and then the television defaults the user intention to the default intention, resulting in a low recognition rate of the actual user intention. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 One of the scenario diagrams of the method for identifying user intention provided in an embodiment of the present application;

[0047] Figure 2 This is one of the structural schematic diagrams of the display device in the method for identifying user intention provided in an embodiment of the present application;

[0048] Figure 3 This is a second structural diagram of a display device in the method for identifying user intention provided in an embodiment of the present application;

[0049] Figure 4 This is a flowchart of a method for identifying user intent provided in an embodiment of the present application;

[0050] Figure 5 This is a second flow chart of the method for identifying user intent provided in an embodiment of the present application;

[0051] Figure 6 A schematic diagram of the intention score distribution of the method for identifying user intention provided in an embodiment of the present application;

[0052] Figure 7 The third flowchart of the method for identifying user intention provided in an embodiment of the present application;

[0053] Figure 8 A schematic diagram of the structure of a text convolutional network model for the method for identifying user intent provided in an embodiment of the present application;

[0054] Figure 9 This is a fourth flow chart of the method for identifying user intent provided in an embodiment of the present application;

[0055] Figure 10 Flowchart 5 of the method for identifying user intent provided in an embodiment of the present application;

[0056] Figure 11 The second scenario diagram of the method for identifying user intention provided in an embodiment of the present application;

[0057] Figure 12 A schematic diagram of the structure of the server provided in the embodiment of the present application;

[0058] Figure 13 One of the schematic diagrams of a chip system provided in an embodiment of the present application;

[0059] Figure 14 A schematic diagram of the structure of a television provided in an embodiment of the present application;

[0060] Figure 15 This is a second schematic diagram of a chip system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the purpose, implementation mode and advantages of the present application clearer, the exemplary implementation mode of the present application will be clearly and completely described below in conjunction with the drawings in the exemplary embodiments of the present application. Obviously, the described exemplary embodiments are only part of the embodiments of the present application, not all of the embodiments.

[0062] Based on the exemplary embodiments described in this application, all other embodiments obtained by those of ordinary skill in the art without making any creative work are within the scope of protection of this application. In addition, although the disclosure in this application is introduced according to one or several exemplary examples, it should be understood that each aspect of these disclosures can also constitute a complete embodiment separately. It should be noted that the brief description of the terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their ordinary and usual meanings.

[0063] Currently, in intelligent voice interaction scenarios, users can control smart home appliances through voice, such as voice assistants like Google Assistant and Samsung Bixby. For example, using a TV as a terminal device, users can control the TV through voice, performing a range of TV control operations such as watching videos, listening to music, and checking the weather. However, in most cases, when a TV receives unclear or ambiguous voice, it cannot determine the user's intent based on the voice. The TV then defaults to the user's intent, resulting in a low recognition rate of the actual user's intent.

[0064] Figure 1 is a schematic diagram of an operation scenario between a display device and a control device according to one or more embodiments of the present application, such as Figure 1 As shown, a user can operate the display device 200 via a mobile terminal 300 and a control device 100. The control device 100 can be a remote controller, and communication between the remote controller and the display device includes infrared protocol communication, Bluetooth protocol communication, wireless or other wired methods to control the display device 200. The user can control the display device 200 by inputting user commands through buttons on the remote controller, voice input, control panel input, etc. In some embodiments, a mobile terminal, tablet computer, computer, laptop computer, and other smart devices can also be used to control the display device 200.

[0065] In some embodiments, the mobile terminal 300 can install software applications with the display device 200, and achieve connection and communication through a network communication protocol to achieve the purpose of one-to-one control operation and data communication. The audio and video content displayed on the mobile terminal 300 can also be transmitted to the display device 200 to achieve a synchronous display function. The display device 200 also communicates data with the server 400 through a variety of communication methods. The display device 200 can be allowed to communicate and connect through a local area network (LAN), a wireless local area network (WLAN) and other networks. The server 400 can provide various content and interactions to the display device 200. The display device 200 can be a liquid crystal display, an OLED display, or a projection display device. In addition to providing a broadcast receiving television function, the display device 200 can also provide an intelligent network TV function that provides computer support functions.

[0066] In some embodiments, the electronic device provided by the embodiments of the present application may be the above-mentioned display device 200. The display device 200 is used to send query information containing voice data to the server 400, so that the server 400 can input the voice data into the dialogFlow model to determine the user's intention. When it is determined that the user's intention is the default intention, the voice data is input into the pre-configured text convolutional network model to determine the intention score of each intention except the default intention; when it is determined that there is an intention with an intention score greater than a first threshold, the first display information containing the intention with an intention score greater than the first threshold is sent to the display device 200.

[0067] Figure 2 FIG. 2 shows a hardware configuration block diagram of a display device 200 according to an exemplary embodiment. Figure 2 The display device 200 shown includes at least one of a tuner and demodulator 210, a communicator 220, a detector 230, an external device interface 240, a controller 250, a display 260, an audio output interface 270, a memory, a power supply, and a user interface 280. The controller includes a central processing unit (CPU), a video processor, an audio processor, a graphics processor, RAM, ROM, and first to nth interfaces for input / output. The display 260 can be at least one of a liquid crystal display (LCD), an OLED display, a touch display, and a projection display, and can also be a projection device and projection screen. The tuner and demodulator 210 receives broadcast television signals via wired or wireless reception, and demodulates audio and video signals, such as EPG data signals, from multiple wireless or wired broadcast television signals. The detector 230 is used to collect signals from the external environment or external interactions. The controller 250 and tuner and demodulator 210 can be located in different separate devices, that is, the tuner and demodulator 210 can also be located in a device external to the main device where the controller 250 is located, such as an external set-top box.

[0068] In some embodiments, the controller 250 controls the operation of the display device and responds to user operations through various software control programs stored in the memory. The controller 250 controls the overall operation of the display device 200. The user can enter user commands through the graphical user interface (GUI) displayed on the display 260, and the user input interface receives the user input commands through the graphical user interface (GUI). Alternatively, the user can enter user commands by inputting specific sounds or gestures, and the user input interface receives the user input commands by recognizing the sounds or gestures through sensors.

[0069] In some embodiments, the sound collector can be a microphone, also known as a "microphone" or "microphone", which is used to convert sound signals into electrical signals. When performing voice interaction, the user can put their mouth close to the microphone to speak and input the sound signal into the microphone. The display device 200 can be provided with at least one microphone. In other embodiments, the display device 200 can be provided with two microphones, which can not only collect sound signals but also realize noise reduction function. In other embodiments, the display device 200 can also be provided with three, four or more microphones to realize sound signal collection, noise reduction, and identification of sound sources, and realize directional recording function, etc.

[0070] Among them, the microphone may be built into the display device 200, or the microphone may be connected to the display device 200 by wire or wireless means. For example, the microphone may be arranged at the lower edge of the display 260 of the display device 200. Of course, the embodiment of the present application does not limit the position of the microphone on the display device 200. Alternatively, the display device 200 may not include a microphone, that is, the above-mentioned microphone is not arranged in the display device 200. The display device 200 may be connected to an external microphone (also referred to as a microphone) through an interface (such as a USB interface 130). The external microphone may be fixed to the display device 200 by an external fixing member (such as a camera bracket with a clip). For example, the external microphone may be fixed to the edge of the display 260 of the display device 200, such as the upper edge, by an external fixing member.

[0071] In some embodiments, a "user interface" is a medium interface for interaction and information exchange between an application or operating system and a user. It enables the conversion between the internal form of information and a form acceptable to the user. A common form of user interface is a graphical user interface (GUI), which refers to a user interface related to computer operations that is displayed graphically. It can be an interface element such as an icon, window, or control displayed on the display of an electronic device. A control can include at least one of the following visual interface elements: an icon, button, menu, tab, text box, dialog box, status bar, navigation bar, widget, etc.

[0072] In some examples, the display device 200 of one or more embodiments is a TV set 1, and the operating system of the TV set 1 is an Android system. Figure 3 As shown, the television 1 can be logically divided into an application layer (abbreviated as “application layer”) 21 , a kernel layer 22 and a hardware layer 23 .

[0073] Among them, Figure 3 As shown, the hardware layer may include Figure 2The controller 250, communicator 220, detector 230, and display 260 are shown. The application layer 21 includes one or more applications. The applications can be system applications or third-party applications. For example, the application layer 21 includes a voice recognition application that can provide a voice interaction interface and services for connecting the television 1 to the server 400.

[0074] The kernel layer 22 serves as a software middleware between the hardware layer and the application layer 21 and is used to manage and control hardware and software resources.

[0075] The server 400 includes a communication control module 201 and an intention recognition module 202. The communication control module 201 is used to establish a communication connection with the TV 1. For example, the speech recognition application in the TV 1 establishes a communication connection with the communication control module 201 of the server 400 by calling the communicator 220.

[0076] In some examples, the kernel layer 22 includes a detector driver, which is used to send the voice data collected by the detector 230 to the voice recognition application. Taking the display device 200 as a TV 1 as an example, when the voice recognition application in the TV 1 is started and the TV 1 establishes a communication connection with the server 400, the detector driver is used to send the voice data of the user input collected by the detector 230 to the voice recognition application. Afterwards, the voice recognition application sends the query information containing the voice data to the intent recognition module 202 in the server 400. The intent recognition module 202 is used to input the voice data sent by the TV 1 into the dialogFlow model. When it is determined that the user intention is the default intention, the voice data is input into a pre-configured text convolutional network model to determine the intention score of each intention except the default intention; the intent recognition module 202 is used to determine that there is an intention with an intention score greater than a first threshold, and then send the first display information containing the intention with an intention score greater than the first threshold to the TV 1.

[0077] The voice data involved in this application may be data authorized by the user or fully authorized by all parties.

[0078] The methods in the following embodiments can all be implemented in the television set 1 having the above hardware structure. In the following embodiments, the methods of the embodiments of the present application are described by taking the above electronic device being the television set 1 as an example.

[0079] The present application provides a method for speech recognition. Figure 4 As shown, the speech recognition method may include S11-S13.

[0080] S11. The server 400 receives query information sent by the TV 1. The query information includes voice data.

[0081] S12. After inputting the voice data into the dialogFlow model and determining that the user intent is the default intent, the server 400 inputs the voice data into a pre-configured text convolutional neural network model (TextConvolutional Neural Networks, TEXTCNN) to determine the intent score of each intent except the default intent.

[0082] In some examples, when a voice assistant using the dialogFlow model receives unrecognizable or semantically ambiguous voice data, the dialogFlow model will default the user's intent to the default intent. For example: the voice data input by the user is "Chinese Paladin", and the actual intention is to watch the TV series "Chinese Paladin". After the voice data "Chinese Paladin" is input, since this sentence pattern is not trained in the training sample, the dialogFlow model directly outputs the default intent, such as the default intent is to query music. At this time, the default intent output by the dialogFlow model will appear, which is not the actual intention of the user. Based on this, an embodiment of the present application provides a speech recognition method, which, when the current dialogFlow model outputs the default intent, uses a pre-configured text convolutional network model to analyze the voice data again, thereby determining the intent score of each intent except the default intent. Afterwards, the user's intent is determined based on the intent score of each intent except the default intent.

[0083] Specifically, the sum of the intent scores for each intent is equal to 1.

[0084] For example, take the example of five intents, namely, opening an application, video search, music playback, weather query, and karaoke. Assuming that the default intent is to open an application, the voice data is input into the dialogFlow model. When it is determined that the user intent is the default intent, the voice data is input into the pre-configured text convolutional network model to determine the intent scores corresponding to video search, music playback, weather query, and karaoke.

[0085] S13. When server 400 determines that an intention with an intent score greater than a first threshold exists, server 400 sends first display information containing the intention with an intent score greater than the first threshold to television 1. The first display information is used to instruct television 1 to execute the intention with an intent score greater than the first threshold and search for voice data.

[0086] In some examples, a higher intent score indicates a higher user interest in the intent. When the intent score is greater than a first threshold, it indicates that the intent of the voice data is clear and the user can directly access the service of the intent.

[0087] Exemplarily, in combination with the example given in S12 above, it is assumed that the intention score corresponding to the video search is 0.95, the intention score for music playback is 0.03, the intention score for weather query is 0.02, and the intention score for karaoke is 0. When the first threshold is 0.9, since the intention score corresponding to the video search is greater than the first threshold. At this time, the server 400 sends the first display information containing the video search to the TV 1. Afterwards, after receiving the first display information, the TV 1 performs a video search and searches for voice data. For example, when the voice data is "Chinese Paladin", the TV 1 performs a video search and searches for "Chinese Paladin", and the display interface of the TV 1 displays the corresponding search results. For example, the search results include the TV series "Chinese Paladin" and the movie "Chinese Paladin".

[0088] In some possible implementation examples, combined with Figure 4 ,like Figure 5 As shown, the speech recognition method provided in the embodiment of the present application further includes: S14-S16.

[0089] S14. The server 400 obtains historical query data corresponding to at least one default intent. The historical query data includes at least one query message, wherein one query message in the at least one query message corresponds to a default intent.

[0090] S15. The server 400 inputs the voice data corresponding to each query information in at least one query information into the text convolutional network model to determine the maximum intent score corresponding to each default intent and the intent corresponding to the maximum intent score.

[0091] In some examples, to better identify user intent, the speech recognition method provided in embodiments of the present application collects speech data corresponding to each time the dialogFlow model determines the user intent as the default intent, thereby obtaining historical query data. Subsequently, each piece of speech data in the historical query data is input into a text convolutional network model to determine the maximum intent score corresponding to each default intent, as well as the intent corresponding to the maximum intent score.

[0092] For example, in combination with the example given in S12 above, taking the historical query data including two voice data, namely voice data 1 and voice data 2, as an example, the steps of determining the maximum intent score corresponding to each default intent and the intent corresponding to the maximum intent score are as follows:

[0093] By inputting voice data 1 into the text convolutional network model, we can determine the intent score for each intent except the default intent. For example, if the default intent is to open an app, we need to determine the intent scores for video search, music playback, weather query, and karaoke. If the intent score for video search is the highest, then the maximum intent score for the default intent corresponding to voice data 1 is determined as the intent score for video search.

[0094] Similarly, by inputting voice data 2 into the text convolutional network model, we can determine the intent score corresponding to each intent except the default intent. For example, if the default intent is video search, we need to determine the intent scores corresponding to opening an app, music playback, weather query, and karaoke. If the intent score corresponding to weather query is the highest, then the maximum intent score for the default intent corresponding to voice data 1 is determined to be the intent score corresponding to weather query.

[0095] S16. The server 400 determines a first threshold and a second threshold according to the maximum intent score corresponding to each default intent and the intent corresponding to the maximum intent score.

[0096] In some examples, since each default intent corresponds to a maximum intent score, and each maximum intent score corresponds to an intent, the total number of intents included in different maximum intent scores can be collected to determine the distribution of the maximum intent scores. In this way, the first threshold and the second threshold can be determined.

[0097] For example, the historical query data contains 26712 voice data. By inputting each voice data in the historical query data into the text convolutional network model, the intent score corresponding to each intent except the default intent can be determined, and the following is obtained: Figure 6 The intent score distribution graph is shown.

[0098] It can be seen that Figure 6The proportion of intentions included in the intention score interval [0.2, 0.9) is high, so the first threshold can be set to 0.9 and the second threshold can be set to 0.2. When the intention score is less than 0.2, it is considered that the intention of the voice data is unclear, and the server 400 cannot determine which intention the voice data belongs to; when the intention score is greater than 0.9, it is considered that the intention of the voice data is clear. At this time, the server 400 can directly execute the intention and search for the voice data. When the intention score is between the intention score interval [0.2, 0.9), these intentions are considered to be relatively clear but it is impossible to determine which intention the user has. At this time, these intentions can be displayed on the terminal display interface so that the user can select the required intention.

[0099] In some possible implementations, combining Figure 4 ,like Figure 5 As shown, the speech recognition method provided in the embodiment of the present application also includes: S17 and S18.

[0100] S17. When the intention scores of all the intentions are less than or equal to the first threshold, the server 400 determines an intention whose intention score is greater than or equal to the second threshold.

[0101] S18. The server 400 sends second display information containing the intention whose intention score is greater than or equal to the second threshold to the TV 1. The second display information is used to instruct the TV 1 to display the intention whose intention score is greater than or equal to the second threshold.

[0102] In some possible implementation examples, combined with Figure 4 ,like Figure 5 As shown, the speech recognition method provided in the embodiment of the present application also includes: S19.

[0103] S19. When the server 400 determines that the intent scores are all less than the second threshold, it sends third display information including the default intent to the TV 1. The third display information is used to instruct the TV 1 to execute the default intent and search for voice data.

[0104] In some examples, when the server determines that the intent scores are all less than the second threshold, it means that the voice data is not clear enough or is ambiguous. At this time, the server 400 cannot continue to judge the user intent, so the server 400 determines that the user intent is the default intent.

[0105] In some feasible examples, the query information also includes an international language code. Figure 4 ,like Figure 5 As shown, the above S12 can be specifically implemented through the following S120 and S121.

[0106] S120: When server 400 inputs the voice data into the dialogFlow model and determines that the user intent is the default intent and the language corresponding to the international language code (languageCode) is the target language, it performs character classification preprocessing on the voice data to determine first classification data. The target language includes one or more of Chinese, Thai, and Japanese.

[0107] In some examples, when the language corresponding to the international language code is Chinese, the server 400 uses Jieba word segmentation to perform character classification preprocessing on the speech data to determine first classification data.

[0108] S121. The server 400 inputs the first classification data into a pre-configured text convolutional network model to determine the intent score of each intent except the default intent.

[0109] In some feasible examples, the query information also includes an international language code. Figure 4 ,like Figure 5 As shown, the above S12 can be specifically implemented through the following S122 and S123.

[0110] S122. When server 400 inputs the voice data into the dialogFlow model and determines that the user intent is the default intent and the language corresponding to the international language code is not the target language, it performs word classification preprocessing on the voice data to determine second classification data. The target language includes one or more of Chinese, Thai, and Japanese.

[0111] In some examples, when the international language code is English, the server 400 uses the Natural Language Toolkit (NLTK) word segmentation to perform word classification preprocessing on the speech data to determine the second classification data.

[0112] S123. The server 400 inputs the second classification data into a pre-configured text convolutional network model to determine the intent score of each intent except the default intent.

[0113] In some possible implementation examples, combined with Figure 4 ,like Figure 7 As shown, the training process of the text convolutional network model in the speech recognition method provided in the embodiment of the present application includes: S20-S23.

[0114] S20: The server 400 obtains a training sentence and a labeling result of the training sentence, wherein the training sentence includes voice data and intent.

[0115] In some examples, to improve the prediction accuracy of a text convolutional network model, the speech recognition method provided in embodiments of the present application includes training sentences that include actual user input speech data and word classification training speech data. The neural network model is trained using the word classification training speech data to obtain a text convolutional network model. The text convolutional network model is then verified using actual user input speech data. This process is repeated repeatedly, resulting in a text convolutional network model with even higher prediction accuracy.

[0116] S21. The server 400 inputs the training sentence into the neural network model.

[0117] S22. The server 400 determines whether the prediction result of the training sentence output by the neural network model matches the labeling result based on the target loss function.

[0118] S23. When the prediction result does not match the labeling result, the server 400 iterates the configuration parameters of the neural network model in a loop until the neural network model converges to obtain a text convolutional network model.

[0119] In some examples, each word in the speech data is composed of an n-dimensional word vector, that is, the matrix size corresponding to the speech data is n×K, where K is the sentence length.

[0120] For example, the speech data for word classification training is "please turn up volume to 15", and the speech data actually input by the user is "please turn up TV sound to 30". As a distinction, the non-numeric segmentation in the speech data for word classification training is marked with "-o", and the numeric segmentation in the speech data for word classification training is marked with "-final.number". No mark is added to the segmentation in the speech data actually input by the user. Figure 8 As shown, the word segmentation of "please turn up volume to 15" is as follows Figure 8 The five participles shown are: "please-o", "turn up-o", "volume-o", "to-o", and "15-final.number". It can be seen that n is equal to 5 and K is equal to 8.

[0121] Segment “please turn up TV sound to 30” and get Figure 8 The six participles shown are: “please”, “turn up”, “TV”, “sound”, “to”, and “30”. It can be seen that n is equal to 6 and K is equal to 8.

[0122] The matrix corresponding to "please turn up volume to 15" and the matrix corresponding to "please turn up TV sound to 30" are input into the embedding layer of the text convolutional network model respectively. Afterwards, the matrix corresponding to "please turn up volume to 15" is convolved with the feature vector N2 in the convolution layer, and the first feature vector after convolution is input into the fully connected layer. Similarly, the matrix corresponding to "please turn up TV sound to 30" is convolved with the convolution kernel in the convolution layer, and the second feature vector after convolution is input into the fully connected layer. Afterwards, the first feature vector and the second feature vector in the fully connected layer are convolved with the convolution kernel to obtain the intent score of each intent. Figure 8 As shown in the figure, the text convolutional network model outputs three intents, namely control.volume.up, control.volume.set, and control.play.play. The intent scores corresponding to control.volume.up, control.volume.set, and control.play.play are 0.4, 0.3, and 0.2, respectively.

[0123] As can be seen, the intention scores corresponding to control.volume.up, control.volume.set, and control.play.play are all less than 0.9 and greater than 0.2. Therefore, server 400 sends the second display information including control.volume.up, control.volume.set, control.play.play, and control.volume.up to TV 1.

[0124] In some possible implementation examples, combined with Figure 4 ,like Figure 9 As shown, the above S12 can be specifically implemented through the following S124 and S125.

[0125] S124. When the server 400 inputs the voice data into the dialogFlow model and determines that the user intention is the default intention, the server 400 inputs the voice data into a pre-configured text convolutional network model to determine the confidence of each intention except the default intention.

[0126] S125. The server 400 determines the intent score corresponding to each intent except the default intent according to the confidence level.

[0127] In some examples, the intent score corresponding to an intent is equal to the confidence level of the intent.

[0128] In some possible implementation examples, combined with Figure 4 ,like Figure 10 As shown, the above S12 can be specifically implemented through the following S124, S126, S127 and -S128.

[0129] S124. When the server 400 inputs the voice data into the dialogFlow model and determines that the user intention is the default intention, the server 400 inputs the voice data into a pre-configured text convolutional network model to determine the confidence of each intention except the default intention.

[0130] S126 . The server 400 obtains the usage frequency of the intent category to which each intent except the default intent belongs.

[0131] In some examples, by collecting voice data sent by all users, the frequency of users' use of different intent categories can be discovered. For example, by collecting user voice data within a preset time period, the intent corresponding to each voice data item can be determined. The total number of intents contained in each intent category can then be determined. The frequency of use corresponding to the intent category can then be determined based on the total number of intents contained in the intent category and the total number of intents contained in the preset time period.

[0132] S127. The server 400 determines the intention coefficient corresponding to each intention category according to the usage frequency.

[0133] In some examples, the usage frequency corresponding to the intent category is equal to the intent coefficient.

[0134] Exemplarily, in combination with the example given in S23 above, assuming that the intent categories include TV control and media resource search, the usage frequency of the intent category is determined as the intent coefficient of the intent category.

[0135] Specifically, the sum of the intent coefficients corresponding to each intent category is equal to 1.

[0136] S128. The server 400 determines the intent score corresponding to each intent except the default intent based on the confidence and the intent coefficient corresponding to each intent category.

[0137] In some examples, the intent score corresponding to an intent is the product of the confidence corresponding to the intent and the intent coefficient corresponding to the intent, and the intent coefficient corresponding to the intent is equal to the intent coefficient corresponding to the intent category to which the intent belongs.

[0138] For example, in combination with the example given in S127 above, assuming that the intention coefficient corresponding to TV control is w1, and the intention coefficient corresponding to media resource search is w2, the intention coefficient attributable to TV control is w1. Figure 1 The confidence level is A, which belongs to the media search intention Figure 2 The confidence level is B. From the above, we can see that Figure 1 The corresponding intention score is A×w1, Figure 2 The corresponding intention score is B×w2.

[0139] The present application provides a method for speech recognition. Figure 5 As shown, the speech recognition method may include S31-S34.

[0140] S31. The TV 1 sends query information to the server 400. The query information includes voice data.

[0141] S32 : The TV 1 receives the first display information sent by the server 400 , which includes the intention whose intention score is greater than the first threshold.

[0142] S33. The TV 1 executes the intention with the intention score greater than the first threshold according to the first display information, and searches for voice data.

[0143] S34 , the TV 1 displays the intention of executing the intention score greater than the first threshold, and searches for the search results of the voice data.

[0144] In some examples, when the search results contain multiple entries, the TV sets will display the entry category corresponding to each entry. For example, in combination with the example given in S13 above, when the voice data is "Chinese Paladin", the TV set 1 performs a video search and searches for "Chinese Paladin". The display of the TV set 1 is as follows: Figure 11 Interface 110 shown in (a) of FIG. Interface 110 includes a title bar 1100 for displaying the search results for "Chinese Paladin," a display bar 1101 for the TV series "Chinese Paladin," and a display bar 1102 for the power supply for "Chinese Paladin." The TV series "Chinese Paladin" displays TV series related to "Chinese Paladin," such as "Chinese Paladin 1" and "Chinese Paladin 2." The movie "Chinese Paladin" displays power supplies related to "Chinese Paladin," such as the power supply for "Chinese Paladin."

[0145] In some examples, such as Figure 5 As shown, the speech recognition method provided in the embodiment of the present application also includes: S35 and S36.

[0146] S35 . The TV 1 receives the second display information sent by the server 400 , which includes the intention whose intention score is greater than or equal to the second threshold.

[0147] S36 . The TV 1 displays the intention having an intention score greater than or equal to the second threshold according to the second display information.

[0148] In some examples, combined with the example given in S12 above, it is assumed that the intentions with an intention score greater than or equal to the second threshold include opening an application, playing music, checking the weather, and singing karaoke. The TV 1 displays the following information according to the second display information: Figure 11 The interface 111 shown in (b) of FIG. The interface 111 includes a title bar 1110 for prompting the user to select the desired intent, a button 1111 for the intent of "playing music", a button 1112 for the intent of "opening an application", a button 1113 for the intent of "inquiring weather", and a button 1114 for the intent of "singing karaoke". In response to the user's selection operation of the button 1111, the TV 1 displays the following Figure 11 The interface 112 shown in (c) of FIG. The interface 112 includes a title bar 1110 for indicating the search result of “Chinese Paladin” and a display bar 1120 for the music of “Chinese Paladin”.

[0149] In some examples, combined Figure 5 ,like Figure 9 As shown, the above S36 can be specifically implemented through the following S360 and S361.

[0150] S360: The TV 1 sorts the intention scores in descending order according to the second display information, and determines a display order for each intention whose intention score is greater than or equal to the second threshold.

[0151] S361. The TV 1 displays intentions having intention scores greater than or equal to a second threshold according to a display order.

[0152] In some examples, such as Figure 5 As shown, the speech recognition method provided in the embodiment of the present application also includes: S37 and S38.

[0153] S37 . The TV 1 receives the third display information including the default intent sent by the server 400 .

[0154] S38. The TV 1 executes the default intention according to the third display information and searches for voice data.

[0155] In some examples, combined with the example given in S36 above, it is assumed that the default intention is music playing. The TV set 1 displays the following information according to the third display information: Figure 11 The interface 113 shown in (d) of FIG. The interface 113 includes a title bar 1110 for indicating the search result of “Chinese Paladin” and an “unrecognizable” button 1130 for prompting the user that the voice data is not recognized.

[0156] The above mainly introduces the solution provided by the embodiment of the present application from the perspective of method. In order to realize the above functions, it includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily appreciate that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0157] In the embodiment of the present application, the server and the television 1 can be divided into functional modules according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical functional division. In actual implementation, there may be other division methods.

[0158] like Figure 12 As shown, an embodiment of the present application provides a structural diagram of a server 400. The server 400 includes a transceiver unit 101 and a processing unit 102.

[0159] A transceiver unit is used to receive query information sent by an electronic device; the query information includes voice data; a processing unit is used to input the voice data received by the receiving unit into a dialogFlow model, and when it is determined that the user intention is a default intention, input the voice data into a pre-configured text convolutional network model to determine the intention score of each intention except the default intention; the processing unit is also used to control the transceiver unit to send a first display information containing the intention score greater than the first threshold to the electronic device when it is determined that there is an intention with an intention score greater than a first threshold, and the first display information is used to instruct the electronic device to execute the intention with an intention score greater than the first threshold and search for voice data.

[0160] In some feasible examples, the transceiver unit is further used to obtain historical query data corresponding to at least one default intent; the historical query data includes at least one query information, and one query information in the at least one query information corresponds to a default intent; the processing unit is further used to input the voice data corresponding to each query information in the at least one query information obtained by the transceiver unit into the text convolutional network model, and determine the maximum intent score corresponding to each default intent, and the intent corresponding to the maximum intent score;

[0161] The processing unit is further configured to determine a first threshold and a second threshold based on a maximum intent score corresponding to each default intent and an intent corresponding to the maximum intent score.

[0162] In some feasible examples, the processing unit is further used to determine an intention whose intention score is greater than or equal to a second threshold when all intention scores are less than or equal to the first threshold; the processing unit is further used to control the transceiver unit to send a second display message containing an intention whose intention score is greater than or equal to the second threshold to the electronic device, and the second display information is used to instruct the electronic device to display the intention whose intention score is greater than or equal to the second threshold.

[0163] In some feasible examples, the processing unit is also used to control the transceiver unit to send a third display message containing the default intent to the electronic device when determining that the intent scores are all less than a second threshold. The third display information is used to instruct the electronic device to execute the default intent and search for voice data.

[0164] In some feasible examples, the query information also includes an international language code; the processing unit is specifically used to perform character classification preprocessing on the voice data to determine first classification data when the international language code received by the transceiver unit is Chinese; the processing unit is specifically used to input the first classification data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent.

[0165] In some feasible examples, the query information also includes an international language code; the processing unit is specifically used to perform word classification preprocessing on the voice data when the international language code received by the transceiver unit is Thai, and determine second classification data; the processing unit is specifically used to input the second classification data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent.

[0166] In some feasible examples, the training process of the text convolutional network model is as follows: the transceiver unit is also used to obtain training sentences and the marking results of the training sentences; wherein the training sentences include voice data and intent; the processing unit is also used to input the training sentences obtained by the transceiver unit into the neural network model; the processing unit is also used to determine whether the prediction results of the training sentences output by the neural network model match the marking results based on the target loss function; the processing unit is also used to cyclically iterate the configuration parameters of the neural network model when the prediction results do not match the marking results until the neural network model converges to obtain the text convolutional network model.

[0167] In some feasible examples, the processing unit is further used to input the voice data received by the transceiver unit into a preconfigured text convolutional network model to determine the confidence of each intent except the default intent; the processing unit is further used to determine the intent score corresponding to each intent except the default intent based on the confidence.

[0168] In some feasible examples, the transceiver unit is further used to obtain the usage frequency of the intent category to which each intent belongs except the default intent; the processing unit is further used to determine the intent coefficient corresponding to each intent category based on the usage frequency obtained by the transceiver unit; the processing unit is further used to input the voice data into a pre-configured text convolutional network model to determine the confidence of each intent except the default intent; the processing unit is further used to determine the intent score corresponding to each intent except the default intent based on the confidence and the intent coefficient corresponding to each intent category.

[0169] Among them, all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and its role will not be repeated here.

[0170] Of course, the server 400 provided in the embodiment of the present application includes but is not limited to the above modules. For example, the server 400 may further include a storage unit 103. The storage unit 103 may be used to store the program code of the write server 400, and may also be used to store data generated by the write server 400 during operation, such as data in a write request.

[0171] As an example, combining Figure 3 The functions implemented by the communication control module 201 in the server 400 are similar to those Figure 12 The functions of the transceiver unit 101 are the same as those of the intention recognition module 202. Figure 12 The function of the processing unit 102 is the same as that of the data storage module 203. Figure 12 The function of the storage unit 103 in is the same.

[0172] This embodiment of the present application further provides a server, which may include a memory and one or more processors. The memory and processors are coupled. The memory is configured to store computer program code, which includes computer instructions. When the processors execute the computer instructions, the server may perform the functions or steps performed by server 400 in the above-described method embodiment.

[0173] The present application also provides a chip system, which can be applied to the server 400 in the above embodiment. Figure 13As shown, the chip system includes at least one processor 1501 and at least one interface circuit 1502. The processor 1501 can be the processor in the above-mentioned server 400. The processor 1501 and the interface circuit 1502 can be interconnected via a line. The processor 1501 can receive and execute computer instructions from the memory of the above-mentioned server 400 through the interface circuit 1502. When the computer instructions are executed by the processor 1501, the server 400 can execute the various steps performed by the server 400 in the above-mentioned embodiment. Of course, the chip system can also include other discrete components, which are not specifically limited in this embodiment of the present application.

[0174] The embodiment of the present application also provides a computer-readable storage medium for storing computer instructions executed by the above-mentioned server 400.

[0175] The embodiment of the present application also provides a computer program product, including computer instructions executed by the above-mentioned server 400.

[0176] like Figure 14 As shown, an embodiment of the present application provides a schematic structural diagram of a television set 1. The television set 1 includes a transceiver unit 301 and a processing unit 302.

[0177] A transceiver unit is used to send query information to a server; the query information includes voice data; the transceiver unit is also used to receive first display information sent by the server containing an intention with an intention score greater than a first threshold; the processing unit is also used to execute the intention with an intention score greater than the first threshold and search for voice data based on the first display information received by the transceiver unit; the display unit is used to display the search results of the processing unit executing the intention with an intention score greater than the first threshold and searching for voice data.

[0178] In some feasible examples, the transceiver unit is further used to receive second display information sent by the server containing an intention whose intention score is greater than or equal to a second threshold; the processing unit is further used to control the display unit to display the intention whose intention score is greater than or equal to the second threshold based on the second display information received by the transceiver unit.

[0179] In some feasible examples, the processing unit is specifically used to sort the intention scores in descending order according to the second display information received by the transceiver unit, and determine the display order of each intention whose intention score is greater than or equal to the second threshold; the processing unit is specifically used to control the display unit to display the intentions whose intention scores are greater than or equal to the second threshold according to the display order.

[0180] In some feasible examples, the transceiver unit is further used to receive a third display message containing a default intent sent by the server; the processing unit is further used to execute the default intent and search for voice data based on the third display message received by the transceiver unit; the display unit is further used to display the search results of the processing unit executing the default intent and searching for voice data.

[0181] Among them, all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, and its role will not be repeated here.

[0182] Of course, the television set 1 provided in the embodiment of the present application includes but is not limited to the above modules. For example, the television set 1 may further include a storage unit 303. The storage unit 303 may be used to store the program code for writing the television set 1, and may also be used to store data generated during the operation of the television set 1, such as data in a write request.

[0183] The present application also provides an electronic device, which may include a memory and one or more processors. The memory and processor are coupled. The memory is used to store computer program code, which includes computer instructions. When the processor executes the computer instructions, the electronic device may perform the functions or steps performed by the electronic device (e.g., television 1) in the above method embodiment.

[0184] The present application also provides a chip system, which can be applied to the television 1 in the above embodiment. Figure 15 As shown, the chip system includes at least one processor 1601 and at least one interface circuit 1602. The processor 1601 can be the processor in the above-mentioned television 1. The processor 1601 and the interface circuit 1602 can be interconnected via a line. The processor 1601 can receive and execute computer instructions from the memory of the above-mentioned television 1 through the interface circuit 1602. When the computer instructions are executed by the processor 1601, the television 1 can execute the various steps performed by the television 1 in the above-mentioned embodiment. Of course, the chip system can also include other discrete components, which are not specifically limited in this embodiment of the present application.

[0185] The embodiment of the present application also provides a computer-readable storage medium for storing computer instructions executed by the television 1 .

[0186] The embodiment of the present application also provides a computer program product, including computer instructions executed by the above-mentioned television 1.

[0187] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0188] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0189] The units described as separate components may or may not be physically separate, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple places. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0190] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0191] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0192] For ease of explanation, the above description has been made in conjunction with specific embodiments. However, the above discussion of some embodiments is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. Based on the above teachings, various modifications and variations can be obtained. The selection and description of the above embodiments are intended to better explain the principles and practical applications, so that those skilled in the art can better use the embodiments and various different variations of the embodiments suitable for specific use considerations.

Claims

1. A speech recognition method, characterized in that: include: Receiving query information sent by an electronic device; The query information includes voice data; When the voice data is input into the dialogFlow model and the user intent is determined to be a default intent, the voice data is input into a pre-configured text convolutional network model to determine an intent score for each intent except the default intent, where a higher intent score indicates a higher degree of user interest in the intent; When it is determined that there is an intention with an intention score greater than a first threshold, a first display information containing the intention with an intention score greater than the first threshold is sent to the electronic device, and the first display information is used to instruct the electronic device to execute the intention with an intention score greater than the first threshold and search for the voice data.

2. The speech recognition method according to claim 1, wherein: The speech recognition method further comprises: Obtain historical query data corresponding to at least one default intent; the historical query data includes at least one query information, and one query information in the at least one query information corresponds to a default intent; Inputting the speech data corresponding to each query information in the at least one query information into the text convolutional network model, and determining the maximum intent score corresponding to each default intent and the intent corresponding to the maximum intent score; The first threshold and the second threshold are determined according to the maximum intent score corresponding to each default intent and the intent corresponding to the maximum intent score.

3. The speech recognition method according to claim 1, wherein: The speech recognition method further comprises: In a case where the intention scores are all less than or equal to the first threshold, determining an intention whose intention score is greater than or equal to a second threshold; Second display information including the intention whose intention score is greater than or equal to a second threshold is sent to the electronic device, where the second display information is used to instruct the electronic device to display the intention whose intention score is greater than or equal to the second threshold.

4. The speech recognition method according to claim 1, wherein: The speech recognition method further comprises: When it is determined that the intention scores are all less than the second threshold, a third display information containing the default intention is sent to the electronic device, and the third display information is used to instruct the electronic device to execute the default intention and search for the voice data.

5. The speech recognition method according to claim 1, wherein: The query information also includes an international language code; Inputting the speech data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent includes: When the language corresponding to the international language code is a target language, performing character classification preprocessing on the speech data to determine first classification data; wherein the target language includes one or more of Chinese, Thai, and Japanese; The first classification data is input into a pre-configured text convolutional network model to determine the intent score of each intent except the default intent.

6. The speech recognition method according to claim 1, wherein: The query information also includes an international language code; Inputting the speech data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent includes: If the language corresponding to the international language code is not the target language, performing word classification preprocessing on the speech data to determine second classification data; wherein the target language includes one or more of Chinese, Thai, and Japanese; The second classification data is input into a pre-configured text convolutional network model to determine the intent score of each intent except the default intent.

7. The speech recognition method according to any one of claims 1 to 6, characterized in that: The training process of the text convolutional network model is as follows: Obtaining a training sentence and a labeling result of the training sentence; wherein the training sentence includes voice data and intent; Inputting the training sentence into the neural network model; Determining, based on a target loss function, whether a prediction result of the training sentence output by the neural network model matches the labeled result; When the prediction result does not match the labeling result, the configuration parameters of the neural network model are iterated cyclically until the neural network model converges to obtain the text convolutional network model.

8. The speech recognition method according to any one of claims 1 to 6, characterized in that: Inputting the speech data into a preconfigured text convolutional network model to determine the intent score of each intent except the default intent includes: Input the speech data into a pre-configured text convolutional network model to determine the confidence of each intent except the default intent; Determine, based on the confidence level, an intent score corresponding to each intent except the default intent.

9. The speech recognition method according to any one of claims 1 to 6, characterized in that: Inputting the speech data into a preconfigured text convolutional network model and determining the intent score of each intent except the default intent includes: Get the usage frequency of the intent category to which each intent belongs except the default intent; Determining an intention coefficient corresponding to each of the intention categories according to the usage frequency; Input the speech data into a pre-configured text convolutional network model to determine the confidence of each intent except the default intent; Determine the intent score corresponding to each intent except the default intent based on the confidence and the intent coefficient corresponding to each intent category.

10. A computer-readable storage medium, characterized in that The method comprises instructions which, when executed on a computer, enable the computer to execute the speech recognition method according to any one of claims 1 to 9.

Images