Silence detection method based on minimum energy wavelet frame

A wavelet framework and silence detection technology, applied in speech analysis, instruments, etc., can solve the problems of high computational complexity of algorithms and limited scope of technology use.

Active Publication Date: 2013-09-25
GUANGZHOU HAIGE COMM GRP INC
View PDF4 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Some traditional silence detection technologies, such as detection technologies based on frequency-domain energy features or spectral entropy, have high detection accuracy, but the computational complexity of the algorithm is high, which greatly limits the scope of use of related technologies

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
  • Silence detection method based on minimum energy wavelet frame
  • Silence detection method based on minimum energy wavelet frame
  • Silence detection method based on minimum energy wavelet frame

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] The silence detection method of the present embodiment is aimed at digital voice signals polluted by Gaussian white noise, based on the following Lemma 1:

[0054] Lemma 1 If the sequence q i [k] : i=1, 2,..., N generates a sum sequence p on the discrete signal space [k] The associated minimum energy wavelet frame, then for any sequence c [j+1，k] , there is a wavelet tower decomposition algorithm as follows:

[0055] c [ j , l ] = 1 2 Σ k p [ k - 2 l ] c [ j + 1 , k ]

[0056] ...

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 discloses a silence detection method based on a minimum energy wavelet frame. The silence detection method comprises the following steps that (1) analog voice signals which are polluted by additive white gaussian noise are sampled to obtain digital voice sampling signals; (2) in a discrete signal space, the minimum energy wavelet frame is used for decomposing the digital voice sampling signals into a plurality of sub signals; (3) framing processing is conducted on all sub signals decomposed in the step (2) simultaneously, the amounts of the frame energy in the same time shafts of all the sub signals are calculated respectively, then the amounts are added, and then the specific values of VAD judgment variables are calculated; (4) a judging device is used for judging the specific values of the VAD judgment variables, the primary judgment result of a three-stage judgment threshold is obtained, a judgment result smoothing algorithm is used, a final VAD judgment result is obtained, and therefore a voice signal silent area and a voice signal acting area in the signals are judged. The silence detection method based on the minimum energy wavelet frame has the advantages of being low in calculating complexity, strong in self-adaptivity, high in VAD effect accuracy and low in hardware cost.

Description

technical field [0001] The invention relates to a silence detection method, in particular to a silence detection method based on a minimum energy wavelet frame, and belongs to the technical field of digital voice signal processing. Background technique [0002] The silence detection (VAD) technology is an important technology of digital voice signal processing. It uses various algorithms to detect the silent time period and the active time period in the conversation, and uses different processing for these two signals according to the detection results. Measures, thereby improving the overall effect of speech signal processing. [0003] Detection accuracy is an important performance index of the VAD algorithm. A conservative VAD algorithm will introduce too many pure noise signals, while an aggressive VAD algorithm will lose part of the voice signal. In addition, the computational complexity and implementation cost of the algorithm are also considerations. Devices are espec...

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): G10L25/78
Inventor 谢映海王健
Owner GUANGZHOU HAIGE COMM GRP INC
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