Unlock instant, AI-driven research and patent intelligence for your innovation.

Voice activity detection method based on two-step permutation entropy

A technology of voice activity detection and permutation entropy, applied in voice analysis, instruments, etc., can solve the problem of high cost

Inactive Publication Date: 2015-12-23
HOHAI UNIV CHANGZHOU
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, a method based on joint audio and video signals can improve the robustness to non-stationary noise, but this method requires additional special sensors and is expensive

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 activity detection method based on two-step permutation entropy
  • Voice activity detection method based on two-step permutation entropy
  • Voice activity detection method based on two-step permutation entropy

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention will be further described below in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

[0032] The present invention relates to a voice activity detection method based on two-step permutation entropy, by calculating the permutation entropy of voice signals, to judge background noise, unvoiced sound and voiced sound. For noisy speech signals, the first step of permutation entropy calculation is performed to identify speech and noise, and the speech signal is separated and then the second step of detection is performed to calculate permutation entropy to obtain the judgment results of unvoiced and voiced sounds.

[0033] Comparison of unvoiced and voiced sound waveforms figure 1 shown. It can be seen from the figure that the voiced sound signal has obvious monotonicity, which decre...

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 voice activity detection method based on two-step permutation entropy, which is characterized by comprising the following steps: (1) framing processing is carried out on a voice signal, and an each-frame voice signal is obtained; (2) short-time energy for the each-frame voice signal is calculated; (3) first-step permutation entropy calculation is carried out on the each-frame voice signal; (4) through the short-time energy and the permutation entropy, first-step judgment is carried out, and noise and the voice signal are detected; (5) second-step permutation entropy calculation is carried out on the detected voice signal; and (6) sonorant and nonsonorant of the voice signal are judged. The method has the beneficial effects that quasi-periodicity between voice signal frame numbers is made full use of, the permutation entropy is used to serve as a voice signal complexity measurement criterion, and the purpose of voice activity detection is realized.

Description

technical field [0001] The invention relates to a voice activity detection method based on two-step permutation entropy, and belongs to the technical field of voice activity detection. Background technique [0002] Voice activity detection technology is to determine the voice signal from a signal containing voice, and plays an important role in voice coding, voice recognition, and voice enhancement. [0003] After years of development, some efficient and practical algorithms have emerged in the field of voice activity detection, such as: short-term energy method, zero-crossing rate method, spectral estimation method, autocorrelation function method, etc. Among them, the detection method represented by the autocorrelation function algorithm has become a recognized standard in this field. However, this type of algorithm also has some disadvantages. For example, when the noise increases, the detection effect decreases. In addition, the above method mainly assumes that the nois...

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): G10L25/84G10L25/93G10L25/21
Inventor 徐宁李海燕鲍静益蒋爱民刘小峰
Owner HOHAI UNIV CHANGZHOU