Speech emotional features measured by power-law distribution based on electroglottography

Lijiang Chen, Xia Mao, Yuli Xue, Mitsuru Ishizuka

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

This study was designed to introduce a kind of novel speech emotional features extracted from Electroglot-tography (EGG). These features were obtained from the power-law distribution coefficient (PLDC) of fundamental frequency (F 0) and duration parameters. First, the segments of silence, voiced and unvoiced (SUV) were distinguished by combining the EGG and speech information. Second, the F 0 of voiced segment and the first-order differential of F 0 was obtained by a cepstrum method. Third, PLDC of voiced segment as well as the pitch rise and pitch down duration were calculated. Simulation results show that the proposed features are closely connected with emotions. Experiments based on Support Vector Machine (SV M) are carried out. The results show that proposed features are better than those commonly used in the case of speaker independent emotion recognition.

Original languageEnglish
Title of host publicationBIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing
Pages131-136
Number of pages6
Publication statusPublished - 2012
Externally publishedYes
EventInternational Conference on Bio-inspired Systems and Signal Processing, BIOSIGNALS 2012 - Vilamoura, Algarve
Duration: 2012 Feb 12012 Feb 4

Other

OtherInternational Conference on Bio-inspired Systems and Signal Processing, BIOSIGNALS 2012
CityVilamoura, Algarve
Period12/2/112/2/4

Fingerprint

Support vector machines
Experiments

Keywords

  • Electroglottography
  • Power-law distribution
  • Speech emotional features

ASJC Scopus subject areas

  • Signal Processing

Cite this

Chen, L., Mao, X., Xue, Y., & Ishizuka, M. (2012). Speech emotional features measured by power-law distribution based on electroglottography. In BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing (pp. 131-136)

Speech emotional features measured by power-law distribution based on electroglottography. / Chen, Lijiang; Mao, Xia; Xue, Yuli; Ishizuka, Mitsuru.

BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing. 2012. p. 131-136.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chen, L, Mao, X, Xue, Y & Ishizuka, M 2012, Speech emotional features measured by power-law distribution based on electroglottography. in BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing. pp. 131-136, International Conference on Bio-inspired Systems and Signal Processing, BIOSIGNALS 2012, Vilamoura, Algarve, 12/2/1.
Chen L, Mao X, Xue Y, Ishizuka M. Speech emotional features measured by power-law distribution based on electroglottography. In BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing. 2012. p. 131-136
Chen, Lijiang ; Mao, Xia ; Xue, Yuli ; Ishizuka, Mitsuru. / Speech emotional features measured by power-law distribution based on electroglottography. BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing. 2012. pp. 131-136
@inproceedings{08a696e2533e442ba9863242f932636d,
title = "Speech emotional features measured by power-law distribution based on electroglottography",
abstract = "This study was designed to introduce a kind of novel speech emotional features extracted from Electroglot-tography (EGG). These features were obtained from the power-law distribution coefficient (PLDC) of fundamental frequency (F 0) and duration parameters. First, the segments of silence, voiced and unvoiced (SUV) were distinguished by combining the EGG and speech information. Second, the F 0 of voiced segment and the first-order differential of F 0 was obtained by a cepstrum method. Third, PLDC of voiced segment as well as the pitch rise and pitch down duration were calculated. Simulation results show that the proposed features are closely connected with emotions. Experiments based on Support Vector Machine (SV M) are carried out. The results show that proposed features are better than those commonly used in the case of speaker independent emotion recognition.",
keywords = "Electroglottography, Power-law distribution, Speech emotional features",
author = "Lijiang Chen and Xia Mao and Yuli Xue and Mitsuru Ishizuka",
year = "2012",
language = "English",
isbn = "9789898425898",
pages = "131--136",
booktitle = "BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing",

}

TY - GEN

T1 - Speech emotional features measured by power-law distribution based on electroglottography

AU - Chen, Lijiang

AU - Mao, Xia

AU - Xue, Yuli

AU - Ishizuka, Mitsuru

PY - 2012

Y1 - 2012

N2 - This study was designed to introduce a kind of novel speech emotional features extracted from Electroglot-tography (EGG). These features were obtained from the power-law distribution coefficient (PLDC) of fundamental frequency (F 0) and duration parameters. First, the segments of silence, voiced and unvoiced (SUV) were distinguished by combining the EGG and speech information. Second, the F 0 of voiced segment and the first-order differential of F 0 was obtained by a cepstrum method. Third, PLDC of voiced segment as well as the pitch rise and pitch down duration were calculated. Simulation results show that the proposed features are closely connected with emotions. Experiments based on Support Vector Machine (SV M) are carried out. The results show that proposed features are better than those commonly used in the case of speaker independent emotion recognition.

AB - This study was designed to introduce a kind of novel speech emotional features extracted from Electroglot-tography (EGG). These features were obtained from the power-law distribution coefficient (PLDC) of fundamental frequency (F 0) and duration parameters. First, the segments of silence, voiced and unvoiced (SUV) were distinguished by combining the EGG and speech information. Second, the F 0 of voiced segment and the first-order differential of F 0 was obtained by a cepstrum method. Third, PLDC of voiced segment as well as the pitch rise and pitch down duration were calculated. Simulation results show that the proposed features are closely connected with emotions. Experiments based on Support Vector Machine (SV M) are carried out. The results show that proposed features are better than those commonly used in the case of speaker independent emotion recognition.

KW - Electroglottography

KW - Power-law distribution

KW - Speech emotional features

UR - http://www.scopus.com/inward/record.url?scp=84861966782&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84861966782&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789898425898

SP - 131

EP - 136

BT - BIOSIGNALS 2012 - Proceedings of the International Conference on Bio-Inspired Systems and Signal Processing

ER -