Voice classification method through using vocal cord modeling inversion

A classification method and voice technology, applied in speech analysis, instruments, etc., can solve problems such as the influence of classification standards, difficulty in simulating glottal waves, etc., and achieve the effect of high accuracy and good inversion effect.

Active Publication Date: 2019-01-01
SUZHOU UNIV
View PDF8 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In actual situations, it is difficult to simulate the glottal wave that matches the actual voice signal by directly sett

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 classification method through using vocal cord modeling inversion
  • Voice classification method through using vocal cord modeling inversion
  • Voice classification method through using vocal cord modeling inversion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0094] Process flow of the present invention such as figure 1 As shown, the glottal wave is first estimated using complex cepstrum phase decomposition (CCPD). Firstly, the pitch period of a frame of voice signal is obtained, and the position of the glottal closure point of a frame of voice signal is obtained through the DYPSA (Dynamic Programming Projected Phase-Slope Algorithm) algorithm, and the position of the glottis closure point corresponds to the pitch period, and each pitch period is obtained The specific position of the closure point of the internal glottis. Obtain the voice signal in each pitch period, use the complex cepstrum method to decompose the voice signal in this period into the maximum phase and minimum phase signals and differentiate them, combine with the position of the glottis closure point, the components of the maximum phase and the glottis The opening coincides, and the smallest phase component coincides with the glottis closing. Before the glottis ...

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 classification method through using vocal cord modeling inversion. All kinds of voices are effectively distinguished from the perspective of a sound production mechanism. The method is characterized by mainly using complex cepstrum phase decomposition to obtain an actual voice glottis wave as a target glottis wave; carrying out vocal cord dynamic model inversion operation through a matching target and a model glottis wave characteristic parameter through using an optimization algorithm; and selecting a normal voice and a special voice to carry out identificationand classification so as to achieve better accuracy. In the invention, after an actual voice signal is input, the actual glottis wave is extracted and is taken as a target, and a genetic algorithm isused to carry out inversion and optimize an original model so as to simulate vocal cord vibration conditions during the sound production of different voices. An experiment result shows that the matched relative error of each characteristic parameter does not exceed 1.95% after model inversion and an inversion effect is good. The normal voice and the special voice are selected to carry out identification and analysis and better accuracy can be achieved.

Description

technical field [0001] The invention relates to the field of voice classification, in particular to a voice classification method based on vocal cord modeling and inversion. Background technique [0002] Voice classification technology is a feature analysis of voice signals to distinguish different types of voices. This technology can be applied to emotional speech analysis and voice quality assessment. The quality of the voice has a direct impact on people's language expression, especially for teachers, announcers and singers. When people speak for a long time or are in a tense and stressful state, the voice will change and even appear hissing sound. [0003] At present, the acoustic analysis method of the voice is widely used, which can only provide the acoustic information of the voice, and cannot be related to the physiological structure of the actual vocal system, and cannot provide a good classification standard. There are large errors in the classification results. ...

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): G10L25/27G10L25/39G10L25/48G10L25/51G10L25/03G10L25/24
CPCG10L25/03G10L25/24G10L25/27G10L25/39G10L25/48G10L25/51
Inventor 孙宝印陶智陈莉媛张晓俊吴迪肖仲喆
Owner SUZHOU UNIV
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