Mandarin emotion recognition combining acoustic and emotional point information

Lijiang Chen, Xia Mao, Pengfei Wei, Yuli Xue, Mitsuru Ishizuka

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

In this contribution, we introduce a novel approach to combine acoustic information and emotional point information for a robust automatic recognition of a speaker's emotion. Six discrete emotional states are recognized in the work. Firstly, a multi-level model for emotion recognition by acoustic features is presented. The derived features are selected by fisher rate to distinguish different types of emotions. Secondly, a novel emotional point model for Mandarin is established by Support Vector Machine and Hidden Markov Model. This model contains 28 emotional syllables which reflect rich emotional information. Finally the acoustic information and emotional point information are integrated by a soft decision strategy. Experimental results show that the application of emotional point information in speech emotion recognition is effective.

Original languageEnglish
Pages (from-to)1-11
Number of pages11
JournalApplied Intelligence
DOIs
Publication statusAccepted/In press - 2012
Externally publishedYes

Fingerprint

Acoustics
Hidden Markov models
Speech recognition
Support vector machines

Keywords

  • Emotional point
  • Fisher rate
  • Hidden Markov model
  • Mandarin emotion recognition
  • Support vector machine

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Mandarin emotion recognition combining acoustic and emotional point information. / Chen, Lijiang; Mao, Xia; Wei, Pengfei; Xue, Yuli; Ishizuka, Mitsuru.

In: Applied Intelligence, 2012, p. 1-11.

Research output: Contribution to journalArticle

Chen, Lijiang ; Mao, Xia ; Wei, Pengfei ; Xue, Yuli ; Ishizuka, Mitsuru. / Mandarin emotion recognition combining acoustic and emotional point information. In: Applied Intelligence. 2012 ; pp. 1-11.
@article{39288a64e791467a954dc1729404eb5f,
title = "Mandarin emotion recognition combining acoustic and emotional point information",
abstract = "In this contribution, we introduce a novel approach to combine acoustic information and emotional point information for a robust automatic recognition of a speaker's emotion. Six discrete emotional states are recognized in the work. Firstly, a multi-level model for emotion recognition by acoustic features is presented. The derived features are selected by fisher rate to distinguish different types of emotions. Secondly, a novel emotional point model for Mandarin is established by Support Vector Machine and Hidden Markov Model. This model contains 28 emotional syllables which reflect rich emotional information. Finally the acoustic information and emotional point information are integrated by a soft decision strategy. Experimental results show that the application of emotional point information in speech emotion recognition is effective.",
keywords = "Emotional point, Fisher rate, Hidden Markov model, Mandarin emotion recognition, Support vector machine",
author = "Lijiang Chen and Xia Mao and Pengfei Wei and Yuli Xue and Mitsuru Ishizuka",
year = "2012",
doi = "10.1007/s10489-012-0352-1",
language = "English",
pages = "1--11",
journal = "Applied Intelligence",
issn = "0924-669X",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Mandarin emotion recognition combining acoustic and emotional point information

AU - Chen, Lijiang

AU - Mao, Xia

AU - Wei, Pengfei

AU - Xue, Yuli

AU - Ishizuka, Mitsuru

PY - 2012

Y1 - 2012

N2 - In this contribution, we introduce a novel approach to combine acoustic information and emotional point information for a robust automatic recognition of a speaker's emotion. Six discrete emotional states are recognized in the work. Firstly, a multi-level model for emotion recognition by acoustic features is presented. The derived features are selected by fisher rate to distinguish different types of emotions. Secondly, a novel emotional point model for Mandarin is established by Support Vector Machine and Hidden Markov Model. This model contains 28 emotional syllables which reflect rich emotional information. Finally the acoustic information and emotional point information are integrated by a soft decision strategy. Experimental results show that the application of emotional point information in speech emotion recognition is effective.

AB - In this contribution, we introduce a novel approach to combine acoustic information and emotional point information for a robust automatic recognition of a speaker's emotion. Six discrete emotional states are recognized in the work. Firstly, a multi-level model for emotion recognition by acoustic features is presented. The derived features are selected by fisher rate to distinguish different types of emotions. Secondly, a novel emotional point model for Mandarin is established by Support Vector Machine and Hidden Markov Model. This model contains 28 emotional syllables which reflect rich emotional information. Finally the acoustic information and emotional point information are integrated by a soft decision strategy. Experimental results show that the application of emotional point information in speech emotion recognition is effective.

KW - Emotional point

KW - Fisher rate

KW - Hidden Markov model

KW - Mandarin emotion recognition

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84860797664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84860797664&partnerID=8YFLogxK

U2 - 10.1007/s10489-012-0352-1

DO - 10.1007/s10489-012-0352-1

M3 - Article

AN - SCOPUS:84860797664

SP - 1

EP - 11

JO - Applied Intelligence

JF - Applied Intelligence

SN - 0924-669X

ER -