Title :
Acoustic Feature Combination for Robust Speech Recognition
Author :
Zolnay, András ; Schlüter, Ralf ; Ney, Hermann
Author_Institution :
Comput. Sci. Dept., Rheinisch-Westfalische Tech. Hochschule, Aachen, Germany
fDate :
March 18-23, 2005
Keywords :
cepstral analysis; feature extraction; matrix algebra; speech processing; speech recognition; vocabulary; German conversational speech; VerbMobil II; acoustic feature combination; automatic speech recognition systems; large-vocabulary task; linear discriminant analysis; log-linear model combination; mel frequency cepstrum coefficients; multiple acoustic features; perceptual linear prediction; robust speech recognition; speech signal; voicedness; Autocorrelation; Automatic speech recognition; Cepstral analysis; Cepstrum; Concatenated codes; Feature extraction; Linear discriminant analysis; Mel frequency cepstral coefficient; Robustness; Speech recognition;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2005. Proceedings. (ICASSP '05). IEEE International Conference on
Print_ISBN :
0-7803-8874-7
DOI :
10.1109/ICASSP.2005.1415149