Acoustic Feature Optimization for Emotion Affected Speech Recognition

Author

Sun, Yanqing ; Zhou, Yu ; Zhao, Qingwei ; Yan, Yonghong

Author_Institution

ThinkIT Speech Lab., Chinese Acad. of Sci., Beijing, China

fYear

2009

fDate

19-20 Dec. 2009

Firstpage

1

Lastpage

4

Abstract

This paper tries to deal with the problem of performance degradation in emotion affected speech recognition. The F-ratio analysis method in statistics is utilized to analyze the significance of different frequency bands for speech unit classification. The result is then used to optimize filter bank design for Mel-frequency cepstral coefficients (MFCC) and perceptual linear prediction (PLP) features respectively in emotion affected speech recognition. Under comparable conditions, the modified features get a relative 40.14% decrease for MFCC and 34.93% for PLP in sentence error rate.

Keywords

acoustic signal processing; emotion recognition; feature extraction; optimisation; speech recognition; statistical analysis; F-ratio analysis method; Mel-frequency cepstral coefficients; acoustic feature optimization; emotion affected speech recognition; filter bank design; perceptual linear prediction; performance degradation; speech unit classification; Acoustics; Algorithm design and analysis; Cepstral analysis; Emotion recognition; Filter bank; Mel frequency cepstral coefficient; Pattern recognition; Speech analysis; Speech recognition; Sun;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Engineering and Computer Science, 2009. ICIECS 2009. International Conference on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-4994-1

Type

conf

DOI

10.1109/ICIECS.2009.5365821

Filename

5365821