BLE-based voice recognition word configuration method, system, device and storage medium

Abstract

This application relates to the field of speech recognition and interaction technology, and in particular to a BLE-based speech recognition word configuration method, system, device, and storage medium. The method includes establishing a short-range wireless connection between a user terminal and a target device and obtaining the target device's current recognition word configuration information; configuring the recognition words based on user input and performing preset rule verification to generate a recognition word configuration data set including wake-up words and command words; distributing the recognition word configuration dataset and performing integrity verification on the target device side; storing the verified recognition word data and completing recognition word adaptation configuration based on the target device's current voice function state; responding to voice input received by the target device, performing wake-up word recognition and command word matching based on the stored recognition word data, and executing the corresponding control operation upon successful matching. This application can improve the controllability of recognition word configuration and the consistency of the speech processing process.

Description

Technical Field

[0001] This application relates to the field of speech recognition and interaction technology, and in particular to a speech recognition word configuration method, system, device and storage medium based on BLE. Background Technology

[0002] With the development of speech recognition and voice interaction technologies, voice wake-up and voice control functions have been widely applied in smart home devices, wearable devices, and various embedded terminals. Users can wake up and control devices through preset voice commands. To ensure the stability of the recognition process and the controllability of device resource usage, existing voice control devices typically pre-store a fixed number of wake-up words and command words in the device firmware. During operation, they match the user's input voice based on a local recognition model and execute the corresponding control command upon successful matching. As user scenarios become increasingly diverse, some devices have introduced customization capabilities on top of the original preset recognition words, allowing users to modify the recognition word content through external terminals, thereby meeting personalized usage needs to a certain extent.

[0003] Currently, existing technologies for configuring recognition words in voice control terminals typically employ either a preset fixed recognition word approach or a limited custom recognition word approach. For the preset fixed recognition word approach, wake-up words and command words are usually pre-written at the factory. During operation, the device uses a built-in recognition word library to recognize and match received voice information, and executes the corresponding control operation upon successful matching. For the limited custom recognition word approach, users typically input custom recognition words via a mobile terminal, which are then transmitted to the device via a network link to update the information, enabling the device to perform voice recognition and response based on the updated recognition word content. In this type of solution, recognition word configuration generally revolves around modifying the wake-up word, while the device performs storage, replacement, and recognition retrieval based on the received recognition word data.

[0004] However, under the aforementioned existing technical solutions, the configuration process of recognition words typically relies on existing mobile terminals and network transmission links. The device side only performs replacement and invocation based on the received recognition word data, lacking a unified constraint mechanism between the configuration method, configuration content, and device-side recognition processing. On the one hand, in the process of making recognition words adjustable, the setting of recognition word content usually does not uniformly limit the length, quantity, and language form, and lacks processing methods for pronunciation differences, making it difficult to establish a stable correspondence between the adjustment of recognition word content and the device-side recognition processing capabilities. On the other hand, the configuration, updating, and invocation of recognition words all depend on existing terminal configuration methods and network transmission paths. When in a usage environment with limited network or unstable connection, the availability and consistency of the recognition word configuration process are difficult to guarantee. At the same time, the invocation process after the recognition word is updated participates in speech processing together with other speech processing functions in the same recognition link, lacking clear processing boundaries, making it easy for different speech processing methods to interact and affect each other. Based on the above processing methods, in application scenarios where the recognition words can be adjusted, it is difficult to form a coordinated and consistent processing relationship between the recognition word configuration method, data transmission method, and device-side recognition and invocation method. As a result, in actual use, the controllability of recognition word configuration, the stability of recognition processing, and the consistency of overall voice interaction cannot be guaranteed at the same time. Summary of the Invention

[0005] This application provides a BLE-based speech recognition word configuration method, system, device, and storage medium, which can improve the controllability of recognition word configuration and the consistency of speech processing. This application provides the following technical solutions: Firstly, this application provides a BLE-based speech recognition word configuration method, the method comprising: Establish a short-range wireless connection between the user terminal and the target device and obtain the target device's current recognition word configuration information in order to enter the recognition word configuration interactive state; Based on user input, the recognition words are configured and processed, and preset rules are validated to generate a set of recognition word configuration data including wake-up words and command words; The set of identification word configuration data is transmitted through the short-range wireless connection, and the integrity of the received identification word data is verified on the target device side. The verified recognition word data is stored and processed in the target device, and the recognition word adaptation configuration is completed based on the current voice function status of the target device. In response to voice input received by the target device, wake-up word recognition and command word matching are performed based on the stored recognition word data, and the corresponding control operation is executed after a successful match.

[0006] In one specific implementation scheme, establishing a short-range wireless connection between the user terminal and the target device and obtaining the target device's current recognition word configuration information includes: The target device enables short-range wireless communication and enters a connectable state; The user terminal searches for target devices that are in a connectable state through the mini-program and displays a list of devices. Based on the user's selection, it initiates a connection request to the target device to establish a communication link. After the connection is established, the user terminal sends a request to the target device to obtain the identification word configuration information through the communication link, and receives the current identification word configuration information returned by the target device.

[0007] In one specific implementation scheme, the target device enables short-range wireless communication and enters a connectable state, including: The target device enables BLE functionality and enters a connectable state, allowing the user terminal to perform BLE scanning and establish a BLE communication link. Alternatively, the target device can simultaneously enable BLE and NFC functions, and the user terminal can trigger the connection establishment with the target device through NFC and establish the communication link through BLE.

[0008] In one specific implementation scheme, the step of configuring the recognition words based on user input and performing preset rule verification to generate a recognition word configuration data set including wake-up words and command words includes: Responding to user input, perform operations such as adding, editing, or deleting recognized words; In the operation of adding or editing recognition words, the input recognition words are validated according to preset rules. The preset rule validation includes at least recognition word type validation, word count validation, language form validation and quantity validation. When a recognized word meets a preset rule, the recognized word is saved to a configuration list, and a set of recognized word configuration data is generated based on the configuration list. When the recognized word does not meet the preset rules, output the prompt information corresponding to the verification result; The identification word configuration data set includes wake-up words and command words.

[0009] In one specific implementation, the step of transmitting the identification word configuration data set via the short-range wireless connection and performing integrity verification on the received identification word data on the target device side includes: In response to the user's confirmation of the configuration list, the user terminal sends the set of identification word configuration data to the target device through the established short-range wireless connection; The target device performs an integrity check on the received recognition word data, and the integrity check includes at least field length check and check code check; When the identified word data fails the integrity check, the target device sends a retransmission request to the user terminal through the short-range wireless connection, and the user terminal performs a retransmission operation based on the retransmission request. When the identified word data passes the integrity check, the target device stores the identified word data in the local non-volatile storage module.

[0010] In one specific implementation scheme, the step of storing and processing the verified recognition word data in the target device and completing the recognition word adaptation configuration based on the current voice function status of the target device includes: Store the verified word data into the target device; Based on the stored recognition word data, custom recognition words are bound to preset device control protocols and audio response messages; Detect whether the target device currently has the voice self-learning function enabled; if the voice self-learning function is detected to be enabled, execute mutual exclusion control to disable the voice self-learning function; After completing the word recognition adaptation configuration, output a prompt sound indicating that the word recognition has taken effect.

[0011] In one specific implementation scheme, the response to voice input received by the target device involves performing wake-up word recognition and command word matching based on stored recognition word data, and executing corresponding control operations upon successful matching, including: In response to the target device receiving voice input, wake-up word recognition is performed on the voice input; After the wake word is successfully recognized, the target device is put into command receiving state; In response to voice input received by the target device in command receiving state, command word matching is performed based on the stored recognition word data; After a command word is successfully matched, the control operation corresponding to the command word is executed.

[0012] Secondly, this application provides a BLE-based speech recognition word configuration system, which adopts the following technical solution: A BLE-based speech recognition word configuration system, comprising: The connection establishment module is used to establish a short-range wireless connection between the user terminal and the target device and obtain the current recognition word configuration information of the target device in order to enter the recognition word configuration interaction state. The recognition word configuration module is used to configure recognition words based on user input and perform preset rule verification to generate a recognition word configuration data set including wake-up words and command words. The data delivery module is used to deliver the identification word configuration data set through the short-range wireless connection, and to perform integrity verification on the received identification word data on the target device side; The device adaptation module is used to store and process the verified recognition word data in the target device, and to complete the recognition word adaptation configuration based on the current voice function status of the target device. The voice control module is used to respond to voice input received by the target device, perform wake-up word recognition and command word matching based on the stored recognition word data, and execute the corresponding control operation after successful matching.

[0013] Thirdly, this application provides an electronic device, the device including a processor and a memory; the memory stores a program, the program being loaded and executed by the processor to implement a BLE-based speech recognition word configuration method as described in the first aspect.

[0014] Fourthly, this application provides a computer-readable storage medium storing a program that, when executed by a processor, is used to implement a BLE-based speech recognition word configuration method as described in the first aspect.

[0015] By establishing a short-range wireless connection between the user terminal and the target device, the current recognition word configuration information of the target device is obtained. Based on the user input, the recognition words are configured and processed, and preset rule verification is performed to generate a recognition word configuration data set. Subsequently, the recognition word configuration data set is sent to the target device through the short-range wireless connection, and integrity verification is performed on the device side. After the verification is passed, the recognition word data is stored and processed, and the recognition word adaptation configuration is completed in combination with the current voice function status of the target device. Finally, when the target device receives voice input, it performs wake-up word recognition and command word matching based on the stored recognition word data and completes the corresponding control operation, thereby forming an integrated processing flow of recognition word configuration, sending, storage and retrieval. The configuration and verification of recognition words are completed in the same process, ensuring that recognition words are uniformly constrained from the generation stage, thus establishing a stable correspondence between the content of the recognition words and the device's recognition processing capabilities. The recognition word configuration data set is distributed via short-range wireless connection and its integrity is verified on the device side, ensuring consistency in the transmission and reception of the recognition word configuration data and avoiding configuration inconsistencies caused by differences in the transmission process. Simultaneously, the device side stores and processes the verified recognition word data and completes adaptation configuration in conjunction with the voice function status, unifying the process of calling recognition word data on the device side with the voice processing process. Based on this, wake-up word recognition and command word matching are performed on the voice input, allowing the recognition word configuration results to directly participate in the voice control process. Therefore, recognition word configuration, data transmission, and device-side recognition calls are coordinated within the same processing link, thereby improving the controllability of recognition word configuration and the consistency of the voice processing process.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, the preferred embodiments of this application are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the BLE-based speech recognition word configuration method in this embodiment of the application.

[0018] Figure 2 This is a schematic diagram of the overall process of the BLE-based speech recognition word configuration method in the embodiments of this application.

[0019] Figure 3 This is a structural block diagram of the BLE-based speech recognition word configuration system in this application embodiment.

[0020] Figure 4 This is a block diagram of an electronic device configured with BLE-based speech recognition words in an embodiment of this application. Detailed Implementation

[0021] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0022] Optionally, this application uses the BLE-based speech recognition word configuration method provided in various embodiments in an electronic device as an example for illustration. The electronic device is a terminal or a server. The terminal can be a computer, tablet computer, etc. This embodiment does not limit the type of electronic device.

[0023] Reference Figure 1 This is a flowchart illustrating a BLE-based speech recognition word configuration method according to an embodiment of this application. The method includes at least the following steps: Step S101: Establish a short-range wireless connection between the user terminal and the target device and obtain the current recognition word configuration information of the target device to enter the recognition word configuration interaction state.

[0024] In step S101, the connection between the user terminal and the target device is established and the recognition word configuration function is activated, enabling the user terminal to interact with the target device and enter the recognition word configuration process.

[0025] Specifically, the target device first enables short-range wireless communication and enters a connectable state. Then, after the user opens the corresponding mini-program on their terminal and enters the recognition word configuration interface, the mini-program automatically searches for nearby connectable target devices and displays the found device information as a device list. Finally, after the user selects the target device from the device list, the mini-program initiates a connection request to establish a communication link between the user's terminal and the target device. If the connection fails during the connection establishment process, the mini-program outputs a connection prompt message and allows the user to re-initiate the connection request to complete the connection establishment. After the connection is established, the user's terminal sends a recognition word configuration information retrieval request to the target device through the communication link. The target device returns its currently stored recognition word configuration information based on this request. The user's terminal receives the current recognition word configuration information returned by the target device and loads it into the mini-program interface, thus entering the recognition word configuration interactive state.

[0026] Furthermore, short-range wireless communication can be implemented using BLE. In this case, the target device enables BLE and enters a connectable state. The user terminal scans for connectable target devices in the vicinity via BLE, and after selecting a target device, initiates a BLE connection request to establish a BLE communication link. Short-range wireless communication can also be implemented using BLE in conjunction with NFC. In this case, the target device simultaneously enables both BLE and NFC. The user terminal triggers a connection with the target device via NFC, then establishes a communication link via BLE and completes the acquisition of identification word configuration information.

[0027] Alternatively, other short-range wireless communication implementation methods can be selected, and this application does not impose any restrictions on the implementation methods of short-range wireless communication.

[0028] In this application, the user terminal preferably uses a mini-program as the entry point for configuring recognition words and establishes a direct communication link with the target device via BLE. During the recognition word configuration process, the recognition word data is directly sent from the user terminal to the target device via the BLE communication link, without the need for relay processing through a server. Based on the above processing method, a direct interaction path is formed between the user terminal and the target device for recognition word configuration and data transmission, thereby enabling the sending and synchronization of recognition word data even in a network-free environment.

[0029] Step S102: Configure the recognition words based on user input and perform preset rule verification to generate a set of recognition word configuration data including wake-up words and command words.

[0030] In step S102, the configuration processing of the recognition words and the generation of the recognition word configuration data set are mainly completed. By performing this step, users can add, edit, and delete recognition words on the mini-program, and perform preset rule verification on the recognition words during the configuration process, so that the recognition words stored in the configuration list constitute a recognition word configuration data set including wake-up words and command words.

[0031] Specifically, after entering the recognition word configuration interface in the mini-program, users can choose to add, edit, or delete words. In the add operation, users choose to add a wake-up word or a command word. The mini-program provides a Chinese input interface to receive the user's input of the recognition word and supports pronunciation selection for polyphonic characters. After the user completes the input, a preset rule verification is triggered. The preset rule verification includes at least recognition word type verification, character count verification, language form verification, and quantity verification. In this embodiment, the wake-up word must be 3 to 6 characters, and the command word must be 2 to 10 characters. Recognition words are input in Chinese, and the configuration list contains a maximum of 1 wake-up word and a maximum of 10 command words. When a recognition word meets the preset rules, it is saved to the mini-program's local configuration list; when a recognition word does not meet the preset rules, the mini-program outputs a prompt message corresponding to the verification result. The prompt message is the reason for the verification failure and is used to instruct the user to modify the input content.

[0032] Furthermore, in the editing operation, the user selects a saved recognition word from the configuration list and enters the editing interface to modify the content of the recognition word or adjust the pronunciation selection of polyphonic characters. After editing, the modified recognition word is re-validated according to preset rules. If the validation passes, the corresponding recognition word in the configuration list is updated; if the validation fails, the corresponding prompt message is output. In the deletion operation, the user selects the target recognition word and confirms the deletion. The mini-program removes the target recognition word from the configuration list and updates the configurable quantity corresponding to the current configuration list. Based on the configuration list processed by the add, edit, and delete operations, a recognition word configuration data set including wake-up words and command words is generated.

[0033] In one optional implementation, the mini-program also supports batch import processing of recognition words. In this case, the user uploads a text file, the mini-program parses the recognition word content in the text file, extracts the recognition word data according to the corresponding formats of the wake word and command word, and then performs preset rule verification on the extracted recognition word data. If the verification passes, the recognition word data is written to the configuration list, and a recognition word configuration data set is generated. The above implementation is suitable for configuration scenarios with a large number of command words.

[0034] In this application, a comprehensive constraint mechanism is established for the configuration of recognition words, encompassing input, editing, saving, and updating. When a user inputs a recognition word, a joint verification is performed, including language form, character count range, and configuration quantity. Furthermore, polyphonic character selection processing is used to adapt the pronunciation of the recognition word. Specifically, language form is used to limit the use of Chinese input for the recognition word; character count range distinguishes the length requirements of wake-up words and command words; configuration quantity limits the number of wake-up words and command words saved in the configuration list; and polyphonic character selection processing determines the corresponding pronunciation for recognition words containing polyphonic characters. Based on this processing method, personalized input during the recognition word configuration process is always within the preset constraints, thereby ensuring that the recognition word content remains relatively consistent with the recognition processing capabilities of the target device.

[0035] Step S103: Send the identification word configuration data set through a short-range wireless connection, and perform an integrity check on the received identification word data on the target device side.

[0036] In step S103, the main tasks are to transmit the identification word configuration data set to the target device and to verify the integrity of the received data on the target device side. By executing this step, the identification word configuration data set generated in step S102 is transmitted to the target device via the established short-range wireless connection, and the target device performs verification processing on the received identification word data.

[0037] Specifically, after the user confirms the recognition word configuration list on the mini-program, a synchronization to the device is triggered. The mini-program sends the recognition word configuration data set to the target device through the established BLE connection. The recognition word configuration data set can be encapsulated in JSON format or a custom binary protocol, and then directly sent to the target device via the BLE communication link. After receiving the recognition word data, the target device performs an integrity check on the received recognition word data. The integrity check includes at least field length verification and checksum verification to determine whether the received recognition word data is complete.

[0038] In implementation, if the identified word data fails the integrity check, the target device sends a retransmission request to the mini-program via BLE connection. Upon receiving the retransmission request, the mini-program retransmits the identified word configuration data set. In one implementation, the mini-program supports up to three retransmissions. If data synchronization is still not completed after the preset number of retransmissions, the mini-program outputs a data synchronization failure message. If the identified word data passes the integrity check, the target device writes the identified word data to its local non-volatile storage module to retain the current identified word configuration data after the device is powered on again.

[0039] Step S104: Store the verified recognition word data in the target device and complete the recognition word adaptation configuration based on the current voice function status of the target device.

[0040] In step S104, the main tasks are to store the verified recognition word data in the target device and configure the recognition word adaptation. By executing this step, the recognition word data that has passed the integrity verification in step S103 is stored on the target device side, and the target device establishes the corresponding speech processing relationship based on the stored recognition word data, thereby entering the recognition word effective state.

[0041] Specifically, after receiving the verified recognition word data in step S103, the target device first stores the recognition word data in the device-side storage area. Then, based on the stored recognition word data, the target device binds the custom recognition words with preset device control protocols and audio response phrases, so that different recognition words correspond to the corresponding device control content and voice response content. For example, the command word "turn on the lights" is bound as a light control command, and the "lights are on" message is set as the corresponding response tone.

[0042] Furthermore, after storing the recognition words, the target device checks the current voice function status to determine if the voice self-learning function is enabled. If the voice self-learning function is enabled, the target device executes mutual exclusion control to disable it, ensuring the current recognition word adaptation configuration takes effect on the target device side. After completing the recognition word adaptation configuration, the target device generates and plays a prompt tone to indicate that the recognition word configuration is complete.

[0043] Step S105: In response to the voice input received by the target device, perform wake-up word recognition and command word matching based on the stored recognition word data, and execute the corresponding control operation after successful matching.

[0044] In step S105, after receiving the user's voice input, the target device recognizes and matches the user's voice content based on the recognition word data that has been adapted and configured in step S104, and executes the control operation corresponding to the recognition word after successful matching.

[0045] Specifically, after receiving voice input from the user during normal operation, the target device first performs wake-up word recognition on the received voice information based on stored recognition word data. When the received voice information successfully matches a stored wake-up word, the target device enters command receiving mode. In command receiving mode, the target device continues to receive voice input from the user and performs command word matching on the received voice information based on locally stored recognition word data. When a command word match is successful, the target device invokes the control command corresponding to that command word to execute the corresponding control operation.

[0046] Furthermore, the control operation can be based on the binding relationship established in step S104 to call the corresponding device control protocol and output the response voice corresponding to the command word. In this way, the target device completes the voice control process from wake-up word recognition, command word matching to control operation execution based on the stored recognition word data.

[0047] Furthermore, preferably, the target device employs a decoder multiplexing approach for the recognition and processing of wake-up words and command words. This means that instead of setting separate decoders for wake-up word recognition and command word recognition on the device side, the same decoder performs the corresponding recognition processing on the stored wake-up words and command words. Based on this processing method, the target device can reduce the decoding resource consumption during the recognition process while completing the recognition word recognition function configuration, thereby reducing the memory usage pressure on the device side.

[0048] In summary, combining Figure 2 This application proposes a method for configuring and distributing speech recognition words. A short-range wireless connection is established between a user terminal and a target device to obtain the target device's current speech recognition word configuration information. Based on this, the user terminal performs configuration processing on the speech recognition words, and verifies the speech recognition words according to preset rules during the configuration process to generate a speech recognition word configuration data set including wake-up words and command words. Then, the speech recognition word configuration data set is distributed to the target device via the short-range wireless connection, and the target device performs integrity verification on the received speech recognition word data. After successful verification, the speech recognition word data is stored in the target device. Further, the target device completes speech recognition word adaptation configuration based on the stored speech recognition word data, establishing a correspondence between the speech recognition words and device-side control content, and performs corresponding processing based on the target device's current voice function status. After speech recognition word configuration is completed, when the target device receives voice input, it performs wake-up word recognition and command word matching based on the stored speech recognition word data, and executes the corresponding control operation after successful matching, thus forming a complete processing chain from speech recognition word configuration, data distribution, device-side adaptation to voice control execution.

[0049] Based on the above technical solutions, it can be seen that this application does not merely replace the content of the recognition words, but rather integrates recognition word configuration, rule verification, data distribution, device-side storage, adaptation configuration, and voice invocation into a unified processing chain: In the recognition word configuration stage, by constraining the length, quantity, language form, and pronunciation adaptation of the recognition words, a stable correspondence is formed between the adjustment range of the recognition word content and the recognition processing capability of the target device, thereby improving the controllability of the recognition word configuration; In the data transmission stage, the recognition word data is directly distributed through a short-range wireless connection, and combined with integrity verification processing, the recognition word configuration and update process no longer depends on the existing network transmission path, thereby improving the availability and consistency of the recognition word configuration process; In the device-side processing stage, by storing and adapting the recognition word data, and processing it in conjunction with the current voice function status of the target device, a clear boundary is formed between the recognition word invocation process and other voice processing functions in the device, thereby reducing the interactive influence between different voice processing methods. Therefore, this application enables a coordinated and consistent processing relationship between the recognition word configuration method, data transmission method, and device-side recognition call method, thereby improving the problems of insufficient controllability of recognition word configuration, low stability of recognition processing, and difficulty in achieving overall consistency of voice interaction in the prior art.

[0050] Figure 3 This is a structural block diagram of a BLE-based speech recognition word configuration system provided in one embodiment of this application. The system includes at least the following modules: The connection establishment module is used to establish a short-range wireless connection between the user terminal and the target device and obtain the current recognition word configuration information of the target device in order to enter the recognition word configuration interaction state. The recognition word configuration module is used to configure recognition words based on user input and perform preset rule verification to generate a recognition word configuration data set including wake-up words and command words. The data delivery module is used to deliver the recognition word configuration data set through a short-range wireless connection and perform integrity verification on the received recognition word data on the target device side; The device adaptation module is used to store and process the verified recognition word data in the target device, and to complete the recognition word adaptation configuration based on the current voice function status of the target device. The voice control module is used to respond to voice input received by the target device, perform wake-up word recognition and command word matching based on the stored recognition word data, and execute the corresponding control operation after successful matching.

[0051] For relevant details, please refer to the above method implementation examples.

[0052] Figure 4This is a block diagram of an electronic device provided in one embodiment of this application. The device includes at least a processor 401 and a memory 402.

[0053] The processor 401 may include one or more processing cores, such as a single-core processor, a dual-core processor, or a multi-core processor. The processor 401 may be implemented using at least one hardware form of CPU, DSP, FPGA, or ASIC, and is used to execute computer program instructions stored in memory 402 to implement the BLE-based speech recognition word configuration method provided in this application embodiment. Specifically, the processor 401 is used to perform the following processes: establishing a short-range wireless connection between the user terminal and the target device and obtaining the current recognition word configuration information of the target device; configuring the recognition words based on user input and performing preset rule verification to generate a recognition word configuration data set including wake-up words and command words; distributing the recognition word configuration data set through the short-range wireless connection and performing integrity verification on the received recognition word data on the target device side; storing the verified recognition word data in the target device and completing recognition word adaptation configuration based on the current voice function state of the target device; and responding to the voice input received by the target device, performing wake-up word recognition and command word matching based on the stored recognition word data, and executing the corresponding control operation after successful matching.

[0054] The memory 402 may include one or more computer-readable storage media, which may be non-transitory. The memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as flash memory, EEPROM or other storage devices, for storing program instructions, identification word configuration information, identification word configuration data set and verified identification word data.

[0055] Those skilled in the art will understand that Figure 4 The structure of the electronic device shown is for illustrative purposes only. The electronic device may include more or fewer components, and this application does not limit this.

[0056] Optionally, this application also provides a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the BLE-based speech recognition word configuration method of the above method embodiments.

[0057] Optionally, this application also provides a computer product including a computer-readable storage medium storing a program that is loaded and executed by a processor to implement the BLE-based speech recognition word configuration method of the above method embodiments.

[0058] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0059] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for configuring words in speech recognition based on BLE, characterized in that, The method includes: Establish a short-range wireless connection between the user terminal and the target device and obtain the target device's current recognition word configuration information in order to enter the recognition word configuration interactive state; Based on user input, the recognition words are configured and processed, and preset rules are validated to generate a set of recognition word configuration data including wake-up words and command words; The set of identification word configuration data is transmitted through the short-range wireless connection, and the integrity of the received identification word data is verified on the target device side. The verified recognition word data is stored and processed in the target device, and the recognition word adaptation configuration is completed based on the current voice function status of the target device. In response to voice input received by the target device, wake-up word recognition and command word matching are performed based on the stored recognition word data, and the corresponding control operation is executed after a successful match.

2. The BLE-based speech recognition word configuration method according to claim 1, characterized in that, The process of establishing a short-range wireless connection between the user terminal and the target device and obtaining the target device's current recognition word configuration information includes: The target device enables short-range wireless communication and enters a connectable state; The user terminal searches for target devices that are in a connectable state through the mini-program and displays a list of devices. Based on the user's selection, it initiates a connection request to the target device to establish a communication link. After the connection is established, the user terminal sends a request to the target device to obtain the identification word configuration information through the communication link, and receives the current identification word configuration information returned by the target device.

3. The BLE-based speech recognition word configuration method according to claim 2, characterized in that, The target device enables short-range wireless communication and enters a connectable state, including: The target device enables BLE functionality and enters a connectable state, allowing the user terminal to perform BLE scanning and establish a BLE communication link. Alternatively, the target device can simultaneously enable BLE and NFC functions, and the user terminal can trigger the connection establishment with the target device through NFC and establish the communication link through BLE.

4. The BLE-based speech recognition word configuration method according to claim 1, characterized in that, The process of configuring and verifying the recognition words based on user input, and then executing preset rule checks, generates a set of recognition word configuration data including wake-up words and command words, including: Responding to user input, perform operations such as adding, editing, or deleting recognized words; In the operation of adding or editing recognition words, the input recognition words are validated according to preset rules. The preset rule validation includes at least recognition word type validation, word count validation, language form validation and quantity validation. When a recognized word meets a preset rule, the recognized word is saved to a configuration list, and a set of recognized word configuration data is generated based on the configuration list. When the recognized word does not meet the preset rules, output the prompt information corresponding to the verification result; The identification word configuration data set includes wake-up words and command words.

5. The BLE-based speech recognition word configuration method according to claim 1, characterized in that, The step of transmitting the identification word configuration data set via the short-range wireless connection and performing integrity verification on the received identification word data on the target device side includes: In response to the user's confirmation of the configuration list, the user terminal sends the set of identification word configuration data to the target device through the established short-range wireless connection; The target device performs an integrity check on the received recognition word data, and the integrity check includes at least field length check and check code check; When the identified word data fails the integrity check, the target device sends a retransmission request to the user terminal through the short-range wireless connection, and the user terminal performs a retransmission operation based on the retransmission request. When the identified word data passes the integrity check, the target device stores the identified word data in the local non-volatile storage module.

6. The BLE-based speech recognition word configuration method according to claim 1, characterized in that, The process of storing and processing the verified recognition word data in the target device, and completing the recognition word adaptation configuration based on the current voice function status of the target device, includes: Store the verified word data into the target device; Based on the stored recognition word data, custom recognition words are bound to preset device control protocols and audio response messages; Detect whether the target device currently has the voice self-learning function enabled; if the voice self-learning function is detected to be enabled, execute mutual exclusion control to disable the voice self-learning function; After completing the word recognition adaptation configuration, output a prompt sound indicating that the word recognition has taken effect.

7. The BLE-based speech recognition word configuration method according to claim 1, characterized in that, The response to voice input received by the target device includes: performing wake-up word recognition and command word matching based on stored recognition word data, and executing corresponding control operations upon successful matching, including: In response to the target device receiving voice input, wake-up word recognition is performed on the voice input; After the wake word is successfully recognized, the target device is put into command receiving state; In response to voice input received by the target device in command receiving state, command word matching is performed based on the stored recognition word data; After a command word is successfully matched, the control operation corresponding to the command word is executed.

8. A BLE-based speech recognition word configuration system, characterized in that, include: The connection establishment module is used to establish a short-range wireless connection between the user terminal and the target device and obtain the current recognition word configuration information of the target device in order to enter the recognition word configuration interaction state. The recognition word configuration module is used to configure recognition words based on user input and perform preset rule verification to generate a recognition word configuration data set including wake-up words and command words. The data delivery module is used to deliver the identification word configuration data set through the short-range wireless connection, and to perform integrity verification on the received identification word data on the target device side; The device adaptation module is used to store and process the verified recognition word data in the target device, and to complete the recognition word adaptation configuration based on the current voice function status of the target device. The voice control module is used to respond to voice input received by the target device, perform wake-up word recognition and command word matching based on the stored recognition word data, and execute the corresponding control operation after successful matching.

9. An electronic device, characterized in that, The device includes a processor and a memory; the memory stores a program that is loaded and executed by the processor to implement a BLE-based speech recognition word configuration method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a program that, when executed by a processor, is used to implement a BLE-based speech recognition word configuration method as described in any one of claims 1 to 7.

Images