Voicing feature integration in SRI's decipher LVCSR system
Authors:
- 28 September 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1 (15206149) , 921
Abstract
We augment the Mel cepstral (MFCC) feature representation with voicing features from an independent front end. The voicing feature front end parameters are optimized for recognition accuracy. The voicing features computed are the normalized autocorrelation peak and a newly proposed entropy of the high-order cepstrum. We explored several alternatives to integrate the voicing features into SRI's DECIPHER system. Promising early results were obtained in a simple system concatenating the voicing features with MFCC features and optimizing the voicing feature window duration. Best results overall came from a more complex system combining a multiframe voicing feature window with the MFCC plus third differential features using linear discriminant analysis and optimizing the number of voicing feature frames. The best integration approach from the single-pass system experiments was implemented in a multi-pass system for large vocabulary testing on the Switchboard database. An average WER reduction of 2% relative was obtained on the NIST Hub-5 dev2001 and eval2002 databases.Keywords
This publication has 6 references indexed in Scilit:
- Extraction methods of voicing feature for robust speech recognition2003
- Use of voicing features in HMM-based speech recognitionSpeech Communication, 2002
- The SuperARV language model2002
- Semi-tied covariance matrices for hidden Markov modelsIEEE Transactions on Speech and Audio Processing, 1999
- Genones: generalized mixture tying in continuous hidden Markov model-based speech recognizersIEEE Transactions on Speech and Audio Processing, 1996
- Large-vocabulary dictation using SRI's DECIPHER speech recognition system: progressive search techniques1993