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

Adaptive voice separating method based on sound source positioning

A technology of speech separation and sound source localization, which is applied in the field of information processing, and can solve the problems of poor robustness, inability to separate and extract speech, and low degree of separation

Active Publication Date: 2018-12-11
NORTHEASTERN UNIV
View PDF2 Cites 20 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the complex and changeable speech environment and the characteristic coupling of multi-speech mixing, the current speech separation technology based on microphone arrays still has low separation degree and poor robustness, and cannot adaptively analyze speech in any sound source environment. The problem of separation and extraction

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
  • Adaptive voice separating method based on sound source positioning
  • Adaptive voice separating method based on sound source positioning
  • Adaptive voice separating method based on sound source positioning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0060] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. The specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0061] An adaptive speech separation method based on sound source localization, the process is as follows figure 1 As shown, the specific method is as follows:

[0062] Step 1: Use a microphone array composed of M microphones to collect the observed environmental audio signal, and confirm the number of environmental sound sources and the direction of arrival of each sound source. The specific steps are as follows:

[0063] Step 1.1: Framing and windowing the voice signals of each channel;

[0064] Step 1.2: Use speech endpoint detection technology to remove audio frames that do not contain speech components by judg...

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 an adaptive voice separating method based on sound source positioning, and relates to the technical field of information processing. The method includes steps: acquiring an audio signal of an observed environment, and confirming the number of sound sources and the direction of arrival of each sound source; generating a dimension reduction matrix P; generating a voice transfer matrix and a delay superposed wave beam coefficient; determining an active sound source of a frequency point and separating voice components; obtaining the obtained voice components and setting non-activated sound source components as zero; and obtaining time domain voice signals of the sound sources. According to the method, the number and the orientation of the sound sources in a current environment can be obtained through a sound source positioning technology, dimension reduction of each frequency band of the voice signal is performed with the cooperation of a PCA whitening technology toobtain an initial separation matrix, frequency components of each sound source channel are separated through the number of the activated sound sources at the frequency point by adaptive usage of the beam forming technology and the FDICA technology to restore the voice components, the obtained signal-to-noise ratio improvement characteristic is higher, better noise suppression performance is achieved, and the method is applicable to any sound source situations in the real voice environment.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to an adaptive speech separation method based on sound source localization. Background technique [0002] In voice systems in complex application environments such as hands-free phones and classrooms, effectively shielding various external signal interference and enhancing voice purity is one of the important issues to improve the performance of the voice system. The use of speech separation technology can effectively extract target speech and remove noise interference, thereby enhancing the signal-to-noise ratio of speech signals. However, due to the complex and changeable speech environment and the characteristic coupling of multi-speech mixing, the current speech separation technology based on microphone arrays still has low separation degree and poor robustness, and cannot adaptively analyze speech in any sound source environment. The problem of separation and ex...

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): G10L21/0216G10L21/0272G10L21/0308
CPCG10L21/0216G10L21/0272G10L21/0308G10L2021/02166
Inventor 王义魏阳杰张克
Owner NORTHEASTERN UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com