Noise-included speech recognition system and method based on monocular camera
A monocular camera and speech recognition technology, which is applied in speech recognition, speech analysis, character and pattern recognition, etc., can solve problems such as speech recognition effect errors, and achieve the effect of improving accuracy
Inactive Publication Date: 2018-05-25
GUANGDONG POLYTECHNIC NORMAL UNIV
View PDF2 Cites 8 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In the above-mentioned invention patents, the effect of speech recognition is improved by processing the audio information, but the noise still participates in the above-mentioned processing process, so there is still a large error in the effect of speech recognition
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0018] The present invention will be further described in detail below in conjunction with the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.
[0019] Such as figure 1 as shown, figure 1 It is a flow chart of a noisy speech recognition system based on a monocular camera of the present invention, wherein the system includes an image acquisition module 10, a visual processing module 20, an audio acquisition module 30, an audio processing module 40, and a speech recognition module 50; The image acquisition module uses a monocular camera to collect the lip shape and outputs it to the visual processing module; the visual processing module processes the lip image and outputs the result to the speech recognition module; the audio acquisition module uses a microphone to collect user audio and outputs it to the audio processing module; the audio processing module processes user audio and outputs it to the speech recogni...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a noise-included speech recognition system and method based on a monocular camera. The system comprises an image collection module, a visual processing module, an audio collection module, an audio processing module and a speech recognition module; the image collection module collects a mouth shape via the monocular camera and outputs the mouth shape to the visual processingmodule; the visual processing module processes a mouth image and outputs a result to the speech recognition module; the audio collection module uses a microphone to collect user audios and outputs the audios to the audio processing module; the audio processing module processes the user audios and outputs the processed audios to the speech recognition module; and the speech recognition module fuses video data with audio data via a data fusion strategy to realize speech recognition. According to the system and method, characteristics of video information and audio information are utilized, thecharacteristic fusion strategy is used, noise interference caused by robot motors and part friction can be avoided effectively, and the accuracy of the speech recognition system is improved.
Description
technical field [0001] The invention relates to speech recognition technology, and specifically designs a noise-bearing speech recognition system and method based on a monocular camera. Background technique [0002] With the development of human-computer interaction technology, robots are expected to have the same perception ability as humans and to work together with humans. To achieve this goal, some researchers use speech technology to make robots understand human speech. [0003] However, a robot in motion will inevitably produce noise, such as the noise generated by electric fans and motors. Because the microphone is closer to the robot, these noises are easier to obtain than the user's language information, resulting in poor speech recognition by the robot. [0004] The invention patent with application publication number CN201610615354.6 discloses a robot control system and control method based on natural language. The method includes receiving natural language sound...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G10L15/22
CPCG10L15/22G06V40/20
Inventor 梁鹏郝刚吴玉婷
Owner GUANGDONG POLYTECHNIC NORMAL UNIV