Voice enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition

A non-negative matrix decomposition and speech enhancement technology, applied in speech analysis, instruments, etc., can solve the problems of poor pure speech quality, poor speech enhancement effect, and inability to obtain multi-frame spectrum speech-specific information, so as to improve speech quality and enhance effect of effect

Inactive Publication Date: 2017-10-13
北京华控智加科技有限公司
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0014] At present, the speech enhancement method based on NMF is still processing the short-term spectrum. This kind of method has the following problems: the speech-specific

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 enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition
  • Voice enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition
  • Voice enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0070] A speech enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition proposed by the present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0071] A kind of speech enhancement method based on multi-frame spectrum and non-negative matrix decomposition proposed by the present invention, the flow chart is as follows figure 2 As shown, the method is divided into three stages: constructing multi-frame spectrum stage, training base matrix stage and speech enhancement stage; including the following steps:

[0072] 1) Construct the multi-frame spectrum stage; specifically include the following steps:

[0073] 1-1) Preprocessing the speech to obtain the short-term frequency spectrum of the speech; preprocessing includes zero-meaning and pre-emphasizing the speech; first performing zero-meaning, subtracting its mean value for the entire speech; then pre-emphasizing: Perf...

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 voice enhancement method and device based on multi-frame spectrum and non-negative matrix decomposition and belongs to the voice enhancement and non-negative matrix decomposition field. The method comprises steps that pure voice, noise and noise-contained voice are pre-processed to acquire short-time spectrum which is converted into multi-frame spectrum; the multi-frame spectrum of the noise and the pure voice is converted into products of corresponding base matrixes and corresponding coefficient matrixes, and a base matrix of the multi-frame spectrum of the noise and a base matrix of the multi-frame spectrum of the pure voice are solved; the two base matrixes are synthesized to form a base matrix of the multi-frame spectrum of the noise-contained voice, the multi-frame spectrum of the noise-contained voice is converted into a product of a base matrix and a coefficient matrix, a coefficient matrix of the multi-frame spectrum of the noise-contained voice is acquired, and an initial estimate of the multi-frame spectrum of the noise and enhanced voice is acquired; through a Wiener filtering method, the multi-frame spectrum of the enhanced voice is acquired and is transformed into a time domain signal, and enhancement voice is lastly acquired. The method is advantaged in that the special voice information is kept, the voice is better reduced, and the voice enhancement effect is improved.

Description

technical field [0001] The invention belongs to the field of speech enhancement and non-negative matrix decomposition, in particular to a speech enhancement method and device based on multi-frame frequency spectrum and non-negative matrix decomposition. Background technique [0002] Speech enhancement, also known as speech noise reduction, is to process noisy speech, remove the noise part in the noisy speech, obtain the pure speech part in the noisy speech, and improve the speech intelligibility while improving the speech quality voice processing technology. Speech enhancement technology can suppress background noise during voice communication and improve communication quality. It can also be used as the preprocessing system of the speech processing system to help the speech processing system resist noise interference and improve the stability of the system. Today, with the rapid development and maturity of electronic information technology, voice enhancement systems are u...

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
IPC IPC(8): G10L21/0216G10L21/0232G10L25/27G10L25/18
CPCG10L21/0216G10L21/0232G10L25/18G10L25/27
Inventor 何亮施梦楠徐灿刘加
Owner 北京华控智加科技有限公司
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