5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Using Missing Feature Theory to Actively Select Features for Robust Speech Recognition with Interruptions, Filtering and Noise KN-37

Richard Lippmann, Beth A. Carlson

Lincoln Laboratory MIT, Lexington, MA, USA

Speech recognizers trained with quiet wide-band speech degrade dramatically with high-pass, low-pass, and notch filtering, with noise, and with interruptions of the speech input. A new and simple approach to compensate for these degradations is presented which uses mel-filter-bank (MFB) magnitudes as input features and missing feature theory to dynamically modify the probability computations performed in Hidden Markov Model recognizers. When the identity of features missing due to filtering or masking is provided, recognition accuracy on a large talker-independent digit recognition task often rises from below 50% to above 95%. These promising results suggest future work to continuously estimate SNR's within MFB bands for dynamic adaptation of speech recognizers.

Full Paper

Bibliographic reference.  Lippmann, Richard / Carlson, Beth A. (1997): "Using missing feature theory to actively select features for robust speech recognition with interruptions, filtering and noise KN-37", In EUROSPEECH-1997, KN37-KN40.