Voice denoising method based on computational auditory scene analysis and countermeasure network model generation

A network model, speech noise reduction technology, applied in speech analysis, instruments, etc., can solve the problems of auditory system damage, high intensity, reducing the accuracy of speech recognition, etc., to achieve the effect of maintaining distortion

Pending Publication Date: 2018-11-13
THE THIRD RES INST OF CHINA ELECTRONICS TECH GRP CORP
View PDF10 Cites 25 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Especially in some professional applications, external noise is unavoidable, and in many cases, the types of noise are complex and the intensity is large
This type of noise will have a serious impact on subsequent speech signal processing, such as reducing the accuracy of speech recognition
In addition, if the voice data containing noise is processed artificially, long-term work will cause damage to the human auditory system

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 computational auditory scene analysis and countermeasure network model generation
  • Voice denoising method based on computational auditory scene analysis and countermeasure network model generation
  • Voice denoising method based on computational auditory scene analysis and countermeasure network model generation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] The present invention will be described in detail below in conjunction with the implementations shown in the drawings, but it should be noted that these implementations are not limitations of the present invention, and those of ordinary skill in the art based on the functions, methods, or structural changes made by these implementations Equivalent transformations or substitutions all fall within the protection scope of the present invention.

[0019] This embodiment provides a speech noise reduction method based on Computational auditory scene analysis (CASA) and Generative adversarial networks (GAN) model, including:

[0020] Step 1, based on the generator (Generator) and discriminator (Discriminator) of the generative confrontation network, the noisy speech is processed to obtain intermediate results;

[0021] Step 2: Process the intermediate results based on the computational auditory scene analysis method to obtain the final result.

[0022] The voice noise reducti...

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 relates to a voice denoising method based on computational auditory scene analysis and countermeasure network model generation. The method comprises the following steps: S1, processing noise-containing noises based on a generator generating a countermeasure network and a discriminator, thus obtaining an intermediate result; and S2, processing the intermediate result based on the computational auditory scene analysis method, thus obtaining a final result. With the method provided by the invention, part of noises in the voice signals obtained under the complicated channel background environment can be removed, and that no distortion occurs on the voice part can be well kept.

Description

technical field [0001] The invention relates to a speech noise reduction method, in particular to a speech noise reduction method based on computational auditory scene analysis and generation of an adversarial network model. Background technique [0002] Speech is the most important means for human beings to transmit information to each other. A piece of speech carries rich information such as the speaker's intention, identity, and emotion. Voice signals can be transmitted through various media such as air, water, and radio. Speech signals are usually interfered by various noises during the propagation process or due to the limitation of acquisition equipment. Especially in some professional applications, external noise is unavoidable, and in many cases, the types of noise are complex and the intensity is relatively large. This type of noise will seriously affect subsequent speech signal processing, such as reducing the accuracy of speech recognition. In addition, if the ...

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): G10L21/0208G10L21/0308
CPCG10L21/0208G10L21/0308
Inventor 陈龙张小博张晓灿
Owner THE THIRD RES INST OF CHINA ELECTRONICS TECH GRP CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap