Method for separating sound of single channels on basis of human sound models

A separation method and single-channel technology, applied in speech analysis, instruments, etc., can solve the problems that cannot meet the performance of the human ear system, and achieve a highly practical effect

Active Publication Date: 2017-05-10
UNISOUND SHANGHAI INTELLIGENT TECH CO LTD +1
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the defects of the prior art, provide a single-channel sound separation method based on the human voice model, and solve the performance problem that the existing separation technology cannot reach the human ear 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
  • Method for separating sound of single channels on basis of human sound models
  • Method for separating sound of single channels on basis of human sound models
  • Method for separating sound of single channels on basis of human sound models

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] The present invention will be further described below in conjunction with specific examples.

[0057] The invention provides a single-channel sound separation method based on the human voice model, which can separate the human voice under extremely noisy conditions, and the quality of the separated human voice signal is high. This single-channel sound separation method can be used for speech enhancement, such as teleconferencing, human-computer interaction scenarios, and can also realize vocal extraction, such as extracting vocals and accompaniment bands from MP3 music signals, and the accompaniment can be used for KTV Singing, the human voice can be used to score the user's singing level. The single-channel sound separation method of the invention has wide application range and strong practicability. The vocal model-based single-channel sound separation method of the present invention will be described below.

[0058] The invention provides a single-channel sound separ...

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 method for separating sound of single channels on the basis of human sound models. The method includes steps of building the human sound models by the aid of source filters; providing input signals; constructing human sound power of the input signals by the aid of vocal cord vibration power models and filter models of vocal tracts and oral cavities; constructing background sound power of the input signals; reconstructing signal power spectra of the input signals according to the constructed human sound power and the constructed background sound power; solving human sound power and background sound power in the reconstructed signal power spectra and separating human sound signals and background sound signals from the input signals according to the solved human sound signals and the solved background sound signals. The built human sound models comprise the vocal cord vibration power models and the filter models of the vocal tracts and the oral cavities. The method for separating the sound of the single channels has the advantages that human sound can be separated from background sound by the aid of the method under extremely noisy conditions, and the method can be used in the fields of speech enhancement, human sound extraction and the like and is high in practicality.

Description

technical field [0001] The invention relates to the technical field of sound separation, in particular to a single-channel sound separation method based on a human voice model. Background technique [0002] Speech and noise separation (Speech and noise separation) is the processing of noisy speech in order to separate the target speech in a complex noise environment. The main goal is to eliminate the interference of environmental noise on speech and improve speech quality. Speech-to-noise separation can be said to be an extension of the Speech Enhancement algorithm, and the noise it processes can even include the voices of other speakers. [0003] With the continuous development of computer signal processing technology and Internet technology, people pay more and more attention to single-channel sound separation technology, which can be used in speech enhancement, human voice extraction and other fields. This single-channel sound separation technology will be used in many c...

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): G10L21/0272
Inventor 曹裕行
Owner UNISOUND SHANGHAI INTELLIGENT TECH CO LTD
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