Enhanced Speech Recognition with Blind Equalization For Robot "WEVER-R2"

Author

Wang, Donglin ; Leung, Henry ; Kwak, Keun-Chang ; Yoon, Hosub

Author_Institution

Univ. of Calgary, Calgary

fYear

2007

fDate

26-29 Aug. 2007

Firstpage

684

Lastpage

688

Abstract

Ambient noise caused by room acoustics usually degrades quality and reliability of speech recognition for the service robot application. In this paper, we propose enhancing the robustness of speech recognition for home service robots by combining blind equalization with classification. In particular, the linear predictive code (LPC) is used for feature extraction. The constant modulus algorithm (CMA) is combined with the radial basis function (RBF) neural network to form a robust classifier with automatic equalization capability to reduce the room service effect. Using real speech data collected by WEVER-R2 acoustic robot, it is shown that the proposed method can increase the speech recognition rate from 74.46% to 87.23% for age recognition and from 80.95% to 95.09% for gender recognition.

Keywords

blind equalisers; feature extraction; linear predictive coding; radial basis function networks; service robots; speech recognition; WEVER-R2 acoustic robot; age recognition; ambient noise; blind equalization; constant modulus algorithm; enhanced speech recognition; feature extraction; gender recognition; linear predictive code; radial basis function neural network; service robot application; Acoustic applications; Automatic speech recognition; Blind equalizers; Degradation; Feature extraction; Linear predictive coding; Neural networks; Noise robustness; Service robots; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Robot and Human interactive Communication, 2007. RO-MAN 2007. The 16th IEEE International Symposium on

Conference_Location

Jeju

Print_ISBN

978-1-4244-1634-9

Electronic_ISBN

978-1-4244-1635-6

Type

conf

DOI

10.1109/ROMAN.2007.4415173

Filename

4415173