Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Voice denoising method based on audio recognition

A voice noise reduction and audio recognition technology, applied in voice analysis, instruments, etc., can solve problems such as increased system complexity, failure to meet requirements, and inability to guarantee filter coordination and complementarity, and achieve the goal of improving integrity and classification accuracy Effect

Inactive Publication Date: 2009-04-08
UNIV OF SCI & TECH BEIJING
View PDF0 Cites 46 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, in the communication voice of manned spaceflight test, due to the complex environment of space and the earth's atmosphere, noise sources are extensive. In addition to background noise, there are also a lot of irregular noise between sentences and words in the conversation, which seriously interferes. normal voice communication
For this type of noise, designing a single filter obviously cannot meet the noise reduction requirements. If a corresponding filter is designed for each possible noise, it will not only greatly increase the complexity of the system, but also cannot guarantee the coordination and complementarity between the filters.
Therefore, for the non-stationary noise environment with frequent noise bursts, the traditional noise reduction methods can no longer meet the requirements, and new noise reduction methods need to be studied

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Voice denoising method based on audio recognition
  • Voice denoising method based on audio recognition
  • Voice denoising method based on audio recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] According to the method described in the above-mentioned summary of the invention, the specific implementation process is as follows:

[0021] 1. Establish a training sample library, read speech and noise samples in sequence, extract audio signal feature parameters, and form a training sample feature vector matrix.

[0022] 2. After acquiring the input signal, divide the original signal into frames with 20ms as a frame, and then process it with a Hamming window. The second step of processing is performed in units of 10 frames, and each step is 5 frames.

[0023] 3. Feature extraction: This method mainly uses Mel cepstrum coefficient (MFCC) and its first-order difference, sub-band energy distribution, and the calculated feature parameters form a one-dimensional feature vector. The specific extraction process is as follows:

[0024] (1) Mel cepstrum coefficient (MFCC) and its first-order difference

[0025] ① Carry out discrete FFT transformation on the input 10-frame ...

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

No PUM Login to View More

Abstract

The invention provides a speech noise reduction method based on audio recognition, which reduces the noise of a receiving end by aiming at the speech communication under complex noise environment, belonging to the field of computer science and technology. Most of the existing noise reduction methods are only suitable for stable noise environment and can not remove the noise under the situations of complex noise environment, especially the situation of frequent mutagenicity noise and the like. The method leads a mode recognition idea in the communication speech noise reduction, divides an audio signal into a speech signal and a non-speech signal, automatically identifies the input signal by extracting the speech characteristic and designing a sorter model, and judges the audio type; if the audio type is noise, the audio signal is removed; if the audio type is speech, the audio signal is remained and processed further. The method meets the real-time requirement and has better reduction noise effect at the same time, can be suitable for the situations with complex communication environments such as manned spaceflight speech communication, construction sites, battlefields and the like, and provides an idea and a method for the noise reduction of signals.

Description

technical field [0001] The present invention proposes a speech noise reduction method—speech noise reduction based on audio recognition, which performs noise reduction processing at the receiving end for speech communication in a complex noise environment. This method introduces the idea of ​​pattern recognition into communication voice noise reduction, which can effectively remove irregular and sudden noise mixed in the voice signal intermittently, and has good adaptability to complex environments, and can be applied to manned spaceflight test voice communication and construction sites , battlefields and other situations where sudden noises are frequent, the environment is complex, and communication channel interference is large, it provides an idea and method for signal noise reduction. Background technique [0002] Wireless voice noise reduction technology refers to the technology of extracting and enhancing useful voice signals from the noise background to reduce noise i...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L21/02G10L21/0272
Inventor 郝红卫高玉峰温博
Owner UNIV OF SCI & TECH BEIJING
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products