Human-computer interaction system based on voice recognition

Abstract

The invention discloses a human-computer interaction system based on voice recognition. The human-computer interaction system comprises a voice unit and an intelligent control unit, the voice unit comprises a voice information acquisition module, a voice information processing module and a voice information recognition module; the voice information acquisition module is used for acquiring user voice signals and encrypting transmitted voice data or processing results so as to protect user privacy and data security, ensure that related laws and regulations and privacy policies are obeyed when voice information is processed, avoid unauthorized access or abuse, and improve user experience. The system can recognize and convert voice input from users with different language backgrounds, the advanced voice recognition module supports recognition of multiple languages and dialects, the system can recognize and convert the voice input from the users with the different language backgrounds by training different language models, high robustness is achieved, and the system is suitable for popularization and application. And different speaking styles, accent and speech speeds can be handled.

Description

Technical Field

[0001] The present invention belongs to the technical field of speech recognition human-computer interaction, and in particular relates to a human-computer interaction system based on speech recognition. Background Art

[0002] Language recognition, also known as Automatic Speech Recognition (ASR), refers to the process of using machines to analyze language signals and make logical judgments to identify language based on the characteristic parameters and grammatical rules of speech units (such as phonemes, syllables or words), and even the regularity of the meaning between speech. Depending on the object of recognition, speech recognition tasks can be roughly divided into three categories: isolated word recognition, keyword recognition, and continuous speech recognition. Among them, continuous speech recognition technology can recognize any continuous speech, such as a sentence or a paragraph, and is the mainstream direction of current research and application.

[0003] After searching, in the prior art, the Chinese patent application number is: CN201911241212.8, a cockpit human-computer interaction system based on artificial intelligence technology, which includes a speech recognition module, a semantic understanding module, an expert database, a speech synthesis module, a decision support module and a pilot model. In this way, core technologies such as microphone array technology, speech recognition technology, speech synthesis technology, semantic understanding technology and intelligent assistance technology (online help) are realized offline through a high-speed embedded processing platform, and integrated into the aircraft cockpit control, so as to systematically realize the intelligent human-computer interaction of the aircraft cockpit.

[0004] The above patents, which collect the user's voice information and use voice recognition technology to identify it, lack the necessary confidentiality protection measures. Among them, the language recognition human-computer interaction system needs to collect the user's voice data for identification and processing. These data may contain sensitive content such as the user's personal information, habits, preferences, etc. Once the system security measures are insufficient or it is attacked by hackers, the user's privacy information may be leaked. Once the user's voice data is leaked, it may be used by criminals for illegal activities such as identity theft, fraud, extortion, etc. In addition, companies may also abuse these data for precision marketing or advertising push, further infringing on the user's privacy rights. Therefore, we need to provide a human-computer interaction system based on voice recognition. Summary of the invention

[0005] The purpose of the present invention is to provide a human-computer interaction system based on speech recognition, which encrypts the transmitted voice data or processing results to protect user privacy and data security, ensures compliance with relevant laws, regulations and privacy policies when processing voice information, avoids unauthorized access or abuse, and the system can recognize and convert voice input from users with different language backgrounds. The advanced speech recognition module supports the recognition of multiple languages ​​and dialects. By training different language models, the system can recognize and convert voice input from users with different language backgrounds, has high robustness, and can cope with different speaking styles, accents and speaking speeds, so as to solve the problem proposed in the above background technology that the language recognition human-computer interaction system in the prior art needs to collect the user's voice data for recognition and processing. These data may contain sensitive content such as the user's personal information, habits, preferences, etc. Once the system security measures are insufficient or attacked by hackers, the user's privacy information may be leaked. Once the user's voice data is leaked, it may be used by criminals for illegal activities such as identity theft, fraud, and extortion.

[0006] To achieve the above object, the present invention adopts the following technical solution: a human-computer interaction system based on speech recognition, comprising: Voice unit and intelligent control unit; The speech unit includes a speech information acquisition module, a speech information processing module and a speech information recognition module; The voice information acquisition module is used to collect user voice signals and convert them into digital signals for analysis by the voice information processing module; The voice information processing module pre-processes the collected voice signals to facilitate recognition by the voice information recognition module; The voice information recognition module is used to convert the processed voice information into text or commands; The intelligent control unit is used to receive and process text instructions from the voice unit.

[0007] Preferably, the voice information collection module is used in conjunction with the voice information processing module, and the voice information collection module receives sound and converts analog voice signals into digital signals.

[0008] Preferably, the speech information processing module includes a process of enhancing the original speech signal to useful information that can be used for subsequent processing, and the speech information processing module includes noise suppression and feature extraction.

[0009] Preferably, the noise suppression is used for noise suppression and feature extraction, and the feature extraction is used to convert the noise-suppressed speech signal into a series of feature vectors that can be used for recognition.

[0010] Preferably, the special extraction includes Mel-frequency cepstral coefficients, linear prediction coefficients and perceptual linear prediction.

[0011] Preferably, the voice information processing module also includes voice coding and parameter coding.

[0012] Preferably, it further comprises a privacy protection unit, and the privacy protection unit is used to encrypt the transmitted voice data or the processing result.

[0013] Preferably, the speech information recognition module also includes a module for training different language models.

[0014] Preferably, the intelligent control unit is responsible for parsing user voice commands, executing decisions, and coordinating with other system components.

[0015] Preferably, the intelligent control unit also includes an adaptive module.

[0016] Technical effects and advantages of the present invention: Compared with the prior art, the human-computer interaction system based on speech recognition proposed by the present invention has the following advantages: The present invention protects user privacy and data security by encrypting the transmitted voice data or processing results, ensures compliance with relevant laws, regulations and privacy policies when processing voice information, avoids unauthorized access or abuse, and the system can recognize and convert voice input from users with different language backgrounds. The advanced voice recognition module supports the recognition of multiple languages ​​and dialects. By training different language models, the system can recognize and convert voice input from users with different language backgrounds, has high robustness, and can cope with different speaking styles, accents and speaking speeds.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a diagram of the human-computer interaction system based on speech recognition of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the accompanying 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, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0020] The present invention provides Figure 1 The human-computer interaction system based on speech recognition shown includes: Voice unit and intelligent control unit; The speech unit includes a speech information acquisition module, a speech information processing module and a speech information recognition module; The voice information acquisition module is used to collect user voice signals and convert them into digital signals for analysis by the voice information processing module; The voice information processing module pre-processes the collected voice signals to facilitate recognition by the voice information recognition module; The voice information recognition module is used to convert the processed voice information into text or commands; The intelligent control unit is used to receive and process text instructions from the voice unit.

[0021] The voice information collection module is used in conjunction with the voice information processing module. The voice information collection module receives the sound and converts the analog voice signal into a digital signal. Specifically, the voice information acquisition module is responsible for collecting the user's voice signal and converting it into a digital signal for subsequent processing. The voice signal emitted by the user is captured by acoustic devices such as microphones. This is the first step in speech recognition and the basis of the entire interaction process. The voice information acquisition module passes the extracted feature vector to the voice processing module for further processing and analysis.

[0022] The speech information processing module includes a process of enhancing the original speech signal to useful information that can be used for subsequent processing, and the speech information processing module includes noise suppression and feature extraction; Furthermore, the impact of background noise and echo on speech signals is reduced or eliminated, the signal-to-noise ratio of speech signals is improved, the start and end points of speech signals are identified, thereby segmenting effective speech segments from continuous audio streams, adjusting the volume of speech signals, ensuring that the signals are within an appropriate range, neither overloaded nor too weak, and improving the spectral characteristics of speech signals by enhancing the energy of the high-frequency part, which is helpful for subsequent feature extraction.

[0023] The noise suppression is used for noise suppression and feature extraction, and the feature extraction is used to convert the noise suppressed speech signal into a series of feature vectors that can be used for recognition.

[0024] The special extraction includes Mel frequency cepstral coefficients (MFCC), linear prediction coefficients (LPC) and perceptual linear prediction (PLP); Specifically, the preprocessed speech signal is converted into a series of feature vectors that can be used for recognition. These feature vectors should be able to effectively represent the key information in the speech signal, such as the identity of the speaker, the speech content, etc. MFCC is the coefficient that makes up the Mel frequency cepstrum, which is derived from the cepstrum of the audio clip. Mel frequency is based on the hearing characteristics of the human ear and has a nonlinear relationship with the Hz frequency. MFCC is the Hz spectrum feature calculated using this relationship; Linear Predictive Coding (LPC) is a widely used technology in speech signal processing, mainly used for compression and parameter extraction of speech signals. Its basic principle is based on an assumption that the current sample point in the speech signal can be expressed by a linear combination of several previous sample points. The basic idea of ​​LPC is to obtain the prediction coefficient of the speech signal by solving the autoregressive model (ARmodel). Specifically, it is assumed that each sample point in the speech signal can be predicted by the linear combination of the p previous sample points, and the coefficient of this linear combination is the LPC coefficient; Perceptual Linear Prediction (PLP) is a speech feature extraction method based on an auditory model. It simulates the human ear's perception of sound to extract speech features that are more in line with human auditory characteristics. PLP converts speech signals into a series of perceptual linear prediction coefficients by analyzing the frequency characteristics of speech signals and the auditory characteristics of the human ear. These coefficients can reflect key features of speech signals such as resonance peaks and vocal tract shapes. At the same time, considering the nonlinear characteristics of the human ear's frequency response and loudness perception of sound, PLP simulates the auditory characteristics of the human ear and extracts speech features that are more in line with human perception habits, which is conducive to improving the accuracy and robustness of speech recognition. Strong anti-noise ability: Since PLP takes into account the nonlinear characteristics of the human ear's frequency response and loudness perception of sound, it can suppress the impact of noise on speech signals to a certain extent and improve the recognizability of speech signals in noisy environments. Moderate computational complexity: Compared with some complex speech feature extraction methods, the computational complexity of PLP is relatively low, which is suitable for real-time speech processing tasks.

[0025] The voice information processing module also includes voice coding and parameter coding; It should be noted that the extracted feature vectors or the original speech signal are converted into a format suitable for storage or transmission. This can include lossless compression and lossy compression techniques to balance speech quality and bandwidth or storage space requirements.

[0026] Parameter Coding: For some applications, such as speech synthesis, it may be necessary to encode the speech signal into a series of parameters, such as fundamental frequency, spectral envelope, etc., so as to reconstruct the speech signal later.

[0027] Also included is a privacy protection unit, the privacy protection unit is used to encrypt the transmitted voice data or processing results; It is worth noting that the transmitted voice data or processing results are encrypted to protect user privacy and data security, ensure compliance with relevant laws, regulations and privacy policies when processing voice information, and avoid unauthorized access or abuse.

[0028] The speech information recognition module also includes a module for training different language models; Furthermore, the advanced speech recognition module supports the recognition of multiple languages ​​and dialects. By training different language models, the system is able to recognize and convert voice input from users with different language backgrounds. The advanced speech recognition module supports the recognition of multiple languages ​​and dialects. By training different language models, the system is able to recognize and convert voice input from users with different language backgrounds. It has high robustness and can cope with different speaking styles, accents and speaking speeds.

[0029] The intelligent control unit is responsible for parsing user voice commands, executing decisions, and coordinating with other system components; Specifically, the intelligent control unit is responsible for receiving the text instructions converted by the voice recognition module and performing in-depth analysis on these instructions to understand the user's real intentions and needs; Decision making: Based on the analysis results, the intelligent control unit will make corresponding decisions to determine how the system should respond to the user's instructions; Execution control: The intelligent control unit sends control signals to other parts of the system (such as hardware devices, software applications, etc.) to perform the operations or tasks required by the user; System coordination: In a complex system, the intelligent control unit also needs to coordinate the work of various subsystems or components to ensure that they can work together to optimize the overall function; The intelligent control unit also includes an adaptive module; It is worth noting that in addition to the current instructions, the intelligent control unit can also combine historical conversations and contextual information to better understand the user's intentions and needs. In addition to the current instructions, the intelligent control unit can also combine historical conversations and contextual information to better understand the user's intentions and needs. Based on a rich knowledge base and rule base, the intelligent control unit can make reasonable decisions and respond to user instructions in the best way.

[0030] In addition, the above-mentioned text creation unit, image generation unit, high-quality image training unit and model optimization processing unit are also used to realize other functions of the above-mentioned human-computer interaction system based on speech recognition when executed, which will not be elaborated here one by one.

[0031] In addition, the present invention also provides a terminal device. The human-computer interaction system based on voice recognition involved in this embodiment is mainly used in the terminal device, which can be a PC, a portable computer, a mobile terminal, or other device with display and processing functions.

[0032] Specifically, the terminal device may include a processor (such as a CPU), a communication bus, a user interface, a network interface, and a memory. The communication bus is used to realize the connection and communication between these components; the user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard); the network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface); the memory may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory, and the memory may optionally be a storage device independent of the aforementioned processor.

[0033] Among them, the memory stores a readable storage medium, and the readable storage medium stores a human-computer interaction program based on voice recognition. The processor can call the human-computer interaction program based on voice recognition stored in the memory and execute the human-computer interaction system based on voice recognition provided by an embodiment of the present invention.

[0034] It is understood that a readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), static random access memories (SRAM), portable compact disk read-only memories (CD-ROM), digital versatile disks (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used here is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0035] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0036] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0037] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A human-computer interaction system based on speech recognition, characterized in that: include: Voice unit and intelligent control unit; The speech unit includes a speech information acquisition module, a speech information processing module and a speech information recognition module; The voice information acquisition module is used to collect user voice signals and convert them into digital signals for analysis by the voice information processing module; The voice information processing module pre-processes the collected voice signals to facilitate recognition by the voice information recognition module; The voice information recognition module is used to convert the processed voice information into text or commands; The intelligent control unit is used to receive and process text instructions from the voice unit.

2. The human-computer interaction system based on speech recognition according to claim 1, characterized in that: The voice information collection module is used in conjunction with the voice information processing module. The voice information collection module receives sound and converts analog voice signals into digital signals.

3. The human-computer interaction system based on speech recognition according to claim 1, characterized in that: The speech information processing module includes a process of enhancing the original speech signal to useful information that can be used for subsequent processing, and the speech information processing module includes noise suppression and feature extraction.

4. The human-computer interaction system based on speech recognition according to claim 3, characterized in that: The noise suppression is used for noise suppression and feature extraction, and the feature extraction is used to convert the noise-suppressed speech signal into a series of feature vectors that can be used for recognition.

5. The human-computer interaction system based on speech recognition according to claim 4, characterized in that: The special extraction includes Mel-frequency cepstral coefficients, linear prediction coefficients and perceptual linear prediction.

6. The human-computer interaction system based on speech recognition according to claim 5, characterized in that: The voice information processing module also includes voice coding and parameter coding.

7. The human-computer interaction system based on speech recognition according to claim 1, characterized in that: It also includes a privacy protection unit, which is used to encrypt the transmitted voice data or processing results.

8. The human-computer interaction system based on speech recognition according to claim 1, characterized in that: The speech information recognition module also includes a module for training different language models.

9. The human-computer interaction system based on speech recognition according to claim 1, characterized in that: The intelligent control unit is responsible for parsing user voice commands, making decisions and executing them, and coordinating with other system components.

10. The human-computer interaction system based on speech recognition according to claim 9, characterized in that: The intelligent control unit also includes an adaptive module.

Images