DocumentCode :
699877
Title :
Missing feature reconstruction and acoustic model adaptation combined for large vocabulary continuous speech recognition
Author :
Remes, Ulpu ; Palomaki, Kalle J. ; Kurimo, Mikko
Author_Institution :
Adaptive Inf. Res. Centre, Helsinki Univ. of Technol., Helsinki, Finland
fYear :
2008
fDate :
25-29 Aug. 2008
Firstpage :
1
Lastpage :
5
Abstract :
Methods for noise robust speech recognition are often evaluated in small vocabulary speech recognition tasks. In this work, we use missing feature reconstruction for noise compensation in large vocabulary continuous speech recognition task with speech data recorded in noisy environments such as cafeterias. In addition, we combine missing feature reconstruction with constrained maximum likelihood linear regression (CMLLR) acoustic model adaptation and propose a new method for finding noise corrupted speech components for the missing feature approach. Using missing feature reconstruction on noisy speech is found to improve the speech recognition performance significantly. The relative error reduction 36% compared to the baseline is comparable to error reductions introduced with acoustic model adaptation, and results further improve when reconstruction and adaptation are used in parallel.
Keywords :
feature extraction; maximum likelihood estimation; regression analysis; signal reconstruction; speech recognition; vocabulary; CMLLR acoustic model adaptation; large vocabulary continuous speech recognition; maximum likelihood linear regression acoustic model adaptation; missing feature reconstruction; noise compensation; relative error reduction; Acoustics; Adaptation models; Hidden Markov models; Noise; Speech; Speech recognition; Vocabulary;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Conference, 2008 16th European
Conference_Location :
Lausanne
ISSN :
2219-5491
Type :
conf
Filename :
7080409
Link To Document :
بازگشت