Speech emotional dimensions region automatic recognition method

An automatic recognition and emotion technology, applied in speech recognition, speech analysis, instruments, etc., can solve problems such as the inability to recognize any emotion effectively, and achieve the effects of robust noise and other interference factors, low cost of misjudgment, and improved recognition performance

Active Publication Date: 2016-05-25
SOUTHEAST UNIV
View PDF6 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a method for automatic recognition of speech emotion dimension regions, which is used to solve the technical problem that existing speech emotion recognition methods cannot effectively recognize any emotion

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
  • Speech emotional dimensions region automatic recognition method
  • Speech emotional dimensions region automatic recognition method
  • Speech emotional dimensions region automatic recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The present invention will be further described below in conjunction with the accompanying drawings.

[0026] (1) Division of emotional regions

[0027] as attached figure 1 As shown, it is a two-dimensional space composed of arousal dimension and valence dimension in speech emotion. Since different speech emotions have different arousal dimensions and valence dimensions, we can classify speech emotion according to the arousal dimension and valence dimension as four regions. Divide it into different categories such as positive and negative. We can further decompose the dimensional space into higher-resolution model regions to accommodate the processing of complex emotions, such as figure 2 shown.

[0028] (2) Extract effective emotional features

[0029] (2-1) First extract the basic acoustic features, including short-term energy, zero-crossing rate, pitch frequency, the first four formant frequencies, and the 12th-order Mel cepstrum coefficient, and perform acous...

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 speech emotional dimensions region automatic recognition method, and belongs to the technical field of speech recognition technology. A characteristic space reconstruction method is adopted to optimize a classifier. Firstly, a basic acoustic characteristic is extracted and optimized to become a standard for distinguishing emotional regions; secondly, the characteristic space reconstruction method is adopted to decompose and pair a plurality of emotional characteristic spaces, an LDA and an PCA module-cascade methods are separately adopted to improve the dispersion degree between target classes; and thirdly, two emotional region partitioning methods that four regions and sixteen regions partitioning methods are provided to perform complex emotion decomposing instead of traditional basic emotional type decomposing, the classifier outputs are combined through the relative calculation to perform the emotional region recognition, so that a better recognition result is obtained.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to an automatic recognition method for a speech emotion dimension area. Background technique [0002] Traditional speech emotion recognition focuses on the analysis of basic emotion categories, such as happiness, anger, surprise, sadness, etc., and it is difficult to effectively identify the dimension of emotion. Traditional sentiment classifiers are mainly used in discrete sentiment models. Before using them, an assumption must be made about the amount and type of target sentiment. At present, in the laboratory environment, successful applications have been obtained, but for many practical emotion recognition applications, it is necessary to automatically recognize specific types of emotions. Traditional methods are generally difficult to estimate what types of emotions may occur, resulting in unrecognized or wrong recognition, and thus cannot process complex human emot...

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): G10L25/63G10L25/15G10L25/24G10L15/08
CPCG10L15/08G10L25/15G10L25/24G10L25/63
Inventor 黄程韦赵力张昕然余华杨晶徐新洲陶华伟
Owner SOUTHEAST 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