Target speech signal enhancement method, system and storage medium based on continuous noise tracking

A target voice and signal enhancement technology, applied in voice analysis, instruments, etc., can solve problems such as limited performance, achieve quality improvement, and reduce noise residual effects

Inactive Publication Date: 2021-01-26
HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the performance of these algorithms is very limited when operating at low SNR especially in non-stationary noise environments

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
  • Target speech signal enhancement method, system and storage medium based on continuous noise tracking
  • Target speech signal enhancement method, system and storage medium based on continuous noise tracking
  • Target speech signal enhancement method, system and storage medium based on continuous noise tracking

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The invention discloses a target speech signal enhancement method based on continuous noise tracking, which can effectively separate the target source signal from the background noise for the noise in life.

[0020] Such as figure 1 As shown, the framework of the present invention consists of two main parts: a speech estimator and a noise tracker.

[0021] Signal model: We consider an additive signal model, y(n)=x(n)+d(n), where y(n) is a noisy speech signal, and x(n) and d(n) represent pure speech signals respectively and noise signal. The relationship in the time-frequency domain is obtained by using the short-time Fourier transform, Y(l,k)=X(l,k)+D(l,k), where l and k represent the frame number and the index of the frequency point, respectively. The expression form of its polar coordinates is: Y=Re jα ,X=Ae jβ and D=Ne jθ . E{|X(l,k)| 2}=λ x and E{|D(l,k)| 2}=λ d are the variances of speech and noise signals, respectively. From figure 1 We see the main fl...

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 present invention provides a target speech signal enhancement method, system and storage medium based on continuous noise tracking. The target speech signal enhancement method includes: Step 1: receiving a noisy speech signal, and performing frame-based windowing processing on the noisy speech signal , use the short-time Fourier transform to obtain the time-frequency domain relationship; step 2: estimate the noise power spectrum; step 3: estimate the speech power spectrum; step 4: estimate the speech signal through the speech estimator; step 5: Inverse Fourier transform, windowing and using overlap-add technique for speech restoration. The beneficial effects of the present invention are: the present invention effectively separates the target voice signal, greatly reduces the residual noise in the voice signal, and greatly improves the quality of the target signal. This is very important for applications such as automatic speech recognition, speaker recognition, human-machine dialogue interface, and hearing aids.

Description

technical field [0001] The invention relates to the technical field of speech processing, in particular to a target speech signal enhancement method, system and storage medium based on continuous noise tracking. Background technique [0002] Noise exists everywhere in life, and the purpose of the speech enhancement algorithm is to improve the quality and intelligibility of the target speech signal polluted by noise. Existing speech enhancement algorithms usually use speech activity detectors to estimate the background noise to achieve target signal enhancement, and these algorithms perform well in stationary noise environments and high SNR conditions. However, the performance of these algorithms is very limited when operating at low SNR especially in non-stationary noise environments. Since the noise in life is more complicated, such as cars, trains passing by, and pedestrians talking and chatting will generate various noises, it is very necessary to develop a speech enhanc...

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 Patents(China)
IPC IPC(8): G10L21/02G10L21/0272G10L25/03G10L25/45
CPCG10L21/02G10L21/0272G10L25/03G10L25/45
Inventor 张啟权王明江陆云韩宇菲张禄孙凤娇
Owner HARBIN INST OF TECH SHENZHEN GRADUATE SCHOOL
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