Artificial intelligence AI voice recognition system
By introducing a voice recognition unit, an interrupt analysis unit and a noise reduction processing unit into the AI voice recognition system, the interference problem during the acquisition of multiple voice question data is solved, and the recognition efficiency and accuracy are improved.
Patent Information
- Application Number
- CN202411001976.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-24
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing AI voice recognition systems are prone to interference when collecting multiple voice question data, resulting in interruption of playback or low recognition efficiency, and external noise reduces recognition accuracy.
An artificial intelligence AI speech recognition system is designed, including a speech recognition unit, an interrupt analysis unit and a noise reduction processing unit. The voice recognition unit processes a plurality of voice commands by reducing the playback volume, the interrupt analysis unit controls the playback process through the password, and the noise reduction processing unit eliminates noise through signal processing and acoustic model.
It effectively enhances the continuity and efficiency of AI speech recognition, interrupts error recognition in a timely manner, and significantly improves recognition accuracy.
Smart Images

Figure CN119091871A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of speech recognition technology, and in particular to an artificial intelligence (AI) speech recognition system. Background Art
[0002] Referring to the Chinese patent, an AI-based intelligent speech recognition method and system (Announcement No.: CN118230722A, Announcement Date: 2024-06-21) This patent processes and analyzes the speech signal to be recognized by utilizing the powerful feature extraction capabilities and structured data processing capabilities of deep learning network models such as convolutional neural networks (CNN) and graph convolutional neural networks (GCN), extracting the local detail waveform features and global contextual semantic association features of the speech signal, and introducing an adaptive attention mechanism to integrate contextual information and visualize the distribution of important features, thereby ensuring the accuracy of semantic recognition.
[0003] The existing AI voice recognition system still has other defects. For example, when the terminal collects voice question data, if multiple voice question data appear, the process of voice command recognition and playback will be disturbed, causing the terminal to interrupt playback or be unable to recognize other voice commands, the continuity of AI voice recognition, and the re-recognized voice question data interferes with the playback process of voice answer data, and the erroneous voice answer data cannot be interrupted in time, wasting more time. The efficiency of AI voice recognition is low, and external noise will also reduce the accuracy of AI voice recognition. For this reason, the present invention proposes an artificial intelligence AI voice recognition system to solve the above-mentioned problems. Summary of the invention
[0004] 1. Technical issues to be resolved
[0005] In view of the deficiencies in the prior art, the present invention provides an artificial intelligence (AI) speech recognition system that solves the problems mentioned in the above background technology.
[0006] (II) Technical solution
[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence AI speech recognition system, comprising:
[0008] Voice recognition unit: The terminal collects basic data of the user and recognizes the basic data so that the terminal generates target data. When a number of the basic data are collected, the terminal analyzes the basic data, wherein the basic data is specifically voice question data and the target data is specifically voice answer data;
[0009] Interruption analysis unit: collects the underlying data of the user, identifies the user according to the underlying data, and when the target data is played, the underlying data identified again interferes with the playing process of the target data, wherein the underlying data is specifically timbre data;
[0010] Noise reduction processing unit: uses signal processing methods to optimize basic data, enhance the signal, and eliminate the noise of the signal.
[0011] Preferably, when the speech recognition unit collects a plurality of the basic data, the terminal analyzes the plurality of the basic data in a specific manner as follows:
[0012] S1: First, the voice question data collected by the terminal is recorded as Yj, where j is the number of voice question data. When the terminal recognizes the voice question data Yj, it will generate corresponding voice answer data Tj. When the number of voice question data Yj is several, the broadcasting process of voice answer data Tj continues;
[0013] S2: Then the voice answer data Tj being broadcast is recorded as ATS, and the volume of ATS playback is SE. When other voice question data Yj are recognized at this time, the volume of SE is reduced to QSE, where QSE≤1 / 2SE. After the other voice question data Yj are recognized, the volume of ATS is restored to SE.
[0014] Preferably, when the voice question data Yj in S1 is recognized and the corresponding voice answer data Tj is generated, a voice database CG is established in the terminal, and a number of voice question data and corresponding voice answer data are set in the voice database CG. When the same data exists in Yj and CG, Tj is extracted and played in CG.
[0015] Preferably, in said S2, when other voice question data Yj are recognized, the corresponding voice answer data Tj are temporarily stored, and after the ATS is played, the next segment is automatically played in the order in which the voice answer data Tj are stored.
[0016] Preferably, when the target data is played in the interruption analysis unit, the specific manner in which the re-identified underlying data interferes with the playing process of the target data is:
[0017] P1: First, the user's first timbre data is collected, recorded as YS, and stored after YS is collected. Then, when the target data is played, if the content played at this time is the content of the recognition error, the password DF is set. After the timbre data YS issues the password DF, if the timbre data YS and the password DF are both correct, the content of the target data playback stops at this time;
[0018] P2: The terminal then prompts whether re-identification is required. After receiving the confirmation password, the terminal re-collects the user's basic data. If a negative password is received, the terminal will continue to play the stopped target data content.
[0019] Preferably, in P2, after obtaining the confirmed password, the terminal re-collects the basic data of the user, and if there is other identified basic data, it is played in sequence.
[0020] Preferably, when the signal is enhanced and the noise of the signal is eliminated in the noise reduction processing unit, the audio speech signal is used as input, the speech signal is converted from the time domain to the frequency domain, and then the feature extraction is handed over to the acoustic model, the acoustic model judges the signal and outputs high-quality signal data.
[0021] Preferably, the acoustic model judges the signal sequence and outputs signal data with good sound quality first. The decoder can calculate the data in the acoustic mode and language model and output the priority word sequence in the recognition result.
[0022] (III) Beneficial effects
[0023] The present invention provides an artificial intelligence (AI) speech recognition system. Compared with the prior art, it has the following beneficial effects:
[0024] The artificial intelligence AI voice recognition system collects several voice question data, and the terminal analyzes the several voice question data. When a voice command is recognized and played, when other commands enter, the playback sound is reduced to enhance the recognition effect of other voice commands, thereby effectively enhancing the continuity of AI voice recognition. The re-recognized voice question data interferes with the playback process of the voice answer data, and the incorrectly recognized voice answer data can be interrupted in time, saving more time and improving the efficiency of AI voice recognition. In addition, the signal is enhanced and the noise of the signal is eliminated, thereby significantly improving the accuracy of AI voice recognition. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] Figure 1 This is a block diagram of an artificial intelligence (AI) speech recognition system according to an embodiment of the present application;
[0026] Figure 2 This is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION
[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0028] See also Figure 1-Figure 2 , the embodiment of the present invention provides four technical solutions:
[0029] Embodiment 1
[0030] An artificial intelligence (AI) speech recognition system, comprising:
[0031] Speech recognition unit: The terminal collects basic data of the user and recognizes the basic data so that the terminal generates target data. When a number of basic data are collected, the terminal analyzes the basic data, where the basic data is specifically voice question data and the target data is specifically voice answer data. When a number of basic data are collected in the speech recognition unit, the terminal analyzes the basic data in the following specific manners:
[0032] S1: First, the voice question data collected by the terminal is recorded as Yj, where j is the number of voice question data. When the terminal recognizes the voice question data Yj, it will generate corresponding voice answer data Tj. When there are several voice question data Yj, the broadcasting process of voice answer data Tj continues. When the voice question data Yj is recognized and the corresponding voice answer data Tj is generated in S1, a voice database CG is established in the terminal, and several voice question data and corresponding voice answer data are set in the voice database CG. When the same data exists in Yj and CG, Tj is extracted and played in CG.
[0033] S2: Then the voice answer data Tj being broadcast is recorded as ATS, and the volume of ATS is SE. When other voice question data Yj is recognized, the volume of SE is reduced to QSE, where QSE≤1 / 2SE. When other voice question data Yj is recognized, the volume of ATS is restored to SE. In S2, when other voice question data Yj is recognized, the corresponding voice answer data Tj is temporarily stored. After ATS is played, the next segment is automatically played according to the storage order of voice answer data Tj.
[0034] Interruption analysis unit: collects the underlying data of the user, identifies the user according to the underlying data, and when playing the target data, the underlying data identified again interferes with the playing process of the target data, wherein the underlying data is specifically timbre data; when playing the target data in the interruption analysis unit, the specific way in which the underlying data identified again interferes with the playing process of the target data is as follows:
[0035] P1: First, the user's first timbre data is collected, recorded as YS, and stored after YS is collected. Then, when the target data is played, if the content played at this time is the content of the recognition error, the password DF is set. After the timbre data YS issues the password DF, if the timbre data YS and the password DF are both correct, the content of the target data playback stops at this time;
[0036] P2: The terminal then prompts whether re-identification is required. After receiving the confirmation password, the terminal re-collects the user's basic data. If a negative password is received, the stopped target data playback content will continue to be played. In P2, after receiving the confirmation password, the terminal re-collects the user's basic data. If there is other recognized basic data, it will be played in order;
[0037] Noise reduction processing unit: Use signal processing to optimize basic data, enhance the signal, and eliminate the noise of the signal. When the signal is enhanced and the noise of the signal is eliminated in the noise reduction processing unit, the audio speech signal is used as input, the speech signal is converted from the time domain to the frequency domain, and then the feature extraction is handed over to the acoustic model. The acoustic model judges the signal and outputs high-quality signal data. The acoustic model judges the signal sequence and outputs signal data with good sound quality first. The decoder can calculate the data in the acoustic mode and language model, and output the priority word sequence in the recognition result.
[0038] Embodiment 2
[0039] An artificial intelligence (AI) speech recognition system, comprising:
[0040] Speech recognition unit: The terminal collects basic data of the user and recognizes the basic data so that the terminal generates target data. When a number of basic data are collected, the terminal analyzes the basic data, where the basic data is specifically voice question data and the target data is specifically voice answer data. When a number of basic data are collected in the speech recognition unit, the terminal analyzes the basic data in the following specific manners:
[0041] S1: First, the voice question data collected by the terminal is recorded as Yj, where j is the number of voice question data. When the terminal recognizes the voice question data Yj, it will generate corresponding voice answer data Tj. When there are several voice question data Yj, the broadcasting process of voice answer data Tj continues. When the voice question data Yj is recognized and the corresponding voice answer data Tj is generated in S1, a voice database CG is established in the terminal, and several voice question data and corresponding voice answer data are set in the voice database CG. When the same data exists in Yj and CG, Tj is extracted and played in CG.
[0042] S2: Then the voice answer data Tj being broadcast is recorded as ATS, and the volume of ATS is SE. When other voice question data Yj is recognized, the volume of SE is reduced to QSE, where QSE≤1 / 2SE. When other voice question data Yj is recognized, the volume of ATS is restored to SE. In S2, when other voice question data Yj is recognized, the corresponding voice answer data Tj is temporarily stored. After ATS is played, the next segment is automatically played according to the storage order of voice answer data Tj.
[0043] Interruption analysis unit: collects the underlying data of the user, identifies the user according to the underlying data, and when playing the target data, the underlying data identified again interferes with the playing process of the target data, wherein the underlying data is specifically timbre data; when playing the target data in the interruption analysis unit, the specific way in which the underlying data identified again interferes with the playing process of the target data is as follows:
[0044] P1: First, the user's first timbre data is collected, recorded as YS, and stored after YS is collected. Then, when the target data is played, if the content played at this time is the content of the recognition error, the password DF is set. After the timbre data YS issues the password DF, if the timbre data YS and the password DF are both correct, the content of the target data playback stops at this time;
[0045] P2: The terminal then prompts whether re-identification is required. After receiving the confirmation password, the terminal re-collects the user's basic data. If a negative password is received, the stopped target data playback content will continue to be played. In P2, after receiving the confirmation password, the terminal re-collects the user's basic data. If there is other recognized basic data, it will be played in order.
[0046] Embodiment 3
[0047] An artificial intelligence (AI) speech recognition system, comprising:
[0048] Voice recognition unit: The terminal collects basic data of the user and recognizes the basic data so that the terminal generates target data. When a number of the basic data are collected, the terminal analyzes the basic data, wherein the basic data is specifically voice question data and the target data is specifically voice answer data;
[0049] Interruption analysis unit: collects the underlying data of the user, identifies the user according to the underlying data, and when the target data is played, the underlying data identified again interferes with the playing process of the target data, wherein the underlying data is specifically timbre data;
[0050] Noise reduction processing unit: uses signal processing methods to optimize basic data, enhance the signal, and eliminate the noise of the signal.
[0051] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.
[0052] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
[0054] Embodiment 4
[0055] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides an electronic device and corresponding embodiments.
[0056] See also Figure 2 , the electronic device 1000 includes a memory 1010 and a processor 1020 .
[0057] The processor 1020 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.
[0058] The memory 1010 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 1020 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 1010 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 1010 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (such as a DVD-ROM, a double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (such as an SD card, a mini SD card, a Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.
[0059] The memory 1010 stores executable codes, and when the executable codes are processed by the processor 1020 , the processor 1020 can execute part or all of the methods described above.
[0060] The scheme of the present application has been described in detail above with reference to the accompanying drawings. In the above embodiments, the description of each embodiment has its own emphasis. For the part that is not described in detail in a certain embodiment, refer to the relevant description of other embodiments. Those skilled in the art should also know that the actions and modules involved in the description are not necessarily required for the present application. In addition, it can be understood that the steps in the method of the embodiment of the present application can be adjusted in order, merged and deleted according to actual needs, and the modules in the device of the embodiment of the present application can be merged, divided and deleted according to actual needs.
[0061] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.
[0062] Alternatively, the present application can also be implemented as a non-temporary machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or electronic device, server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.
[0063] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the application herein may be implemented as electronic hardware, computer software, or combinations of both.
[0064] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system and method according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, a program segment or a code, and the part of the module, the program segment or the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or the flow chart, and the combination of the square boxes in the block diagram and / or the flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0065] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. An artificial intelligence (AI) speech recognition system, characterized in that: include: Voice recognition unit: The terminal collects basic data of the user and recognizes the basic data so that the terminal generates target data. When a number of the basic data are collected, the terminal analyzes the basic data, wherein the basic data is specifically voice question data and the target data is specifically voice answer data; Interruption analysis unit: collects the underlying data of the user, identifies the user according to the underlying data, and when the target data is played, the underlying data identified again interferes with the playing process of the target data, wherein the underlying data is specifically timbre data; Noise reduction processing unit: uses signal processing methods to optimize basic data, enhance the signal, and eliminate the noise of the signal.
2. An artificial intelligence (AI) speech recognition system according to claim 1, characterized in that: When the speech recognition unit collects a plurality of the basic data, the terminal analyzes the plurality of the basic data in a specific manner as follows: S1: First, the voice question data collected by the terminal is recorded as Yj, where j is the number of voice question data. When the terminal recognizes the voice question data Yj, it will generate corresponding voice answer data Tj. When the number of voice question data Yj is several, the broadcasting process of voice answer data Tj continues; S2: Then the voice answer data Tj being broadcast is recorded as ATS, and the volume of ATS playback is SE. When other voice question data Yj are recognized at this time, the volume of SE is reduced to QSE, where QSE≤1 / 2SE. After the other voice question data Yj are recognized, the volume of ATS is restored to SE.
3. The artificial intelligence (AI) speech recognition system according to claim 1, characterized in that: When the voice question data Yj in S1 is recognized and the corresponding voice answer data Tj is generated, a voice database CG is established in the terminal, and a number of voice question data and corresponding voice answer data are set in the voice database CG. When the same data exists in Yj and CG, Tj is extracted and played in CG.
4. The artificial intelligence (AI) speech recognition system according to claim 1, characterized in that: In the above S2, when other voice question data Yj are recognized, the corresponding voice answer data Tj are temporarily stored. After the ATS is played, the next segment is automatically played according to the storage order of the voice answer data Tj.
5. The artificial intelligence (AI) speech recognition system according to claim 1, characterized in that: When the target data is played in the interrupt analysis unit, the specific manner in which the re-identified underlying data interferes with the playing process of the target data is as follows: P1: First, the user's first timbre data is collected, recorded as YS, and stored after YS is collected. Then, when the target data is played, if the content played at this time is the content of the recognition error, the password DF is set. After the timbre data YS issues the password DF, if the timbre data YS and the password DF are both correct, the content of the target data playback stops at this time; P2: The terminal then prompts whether re-identification is required. After receiving the confirmation password, the terminal re-collects the user's basic data. If a negative password is received, the terminal will continue to play the stopped target data content.
6. An artificial intelligence (AI) speech recognition system according to claim 5, characterized in that: In P2, after obtaining the confirmed password, the terminal re-collects the user's basic data. If there is other recognized basic data, it is played in sequence.
7. The artificial intelligence (AI) speech recognition system according to claim 1, characterized in that: When the noise reduction processing unit enhances the signal and eliminates the noise of the signal, the audio speech signal is used as input, the speech signal is converted from the time domain to the frequency domain, and then the feature extraction is handed over to the acoustic model. The acoustic model judges the signal and outputs high-quality signal data.
8. An artificial intelligence (AI) speech recognition system according to claim 7, characterized in that: The acoustic model is used to judge the signal sequence and output the signal data with good sound quality first. The decoder can calculate the data in the acoustic mode and language model and output the priority word sequence in the recognition result.
Citation Information
Cited By
Method and device for voice response
WO2026212566A1
