7th International Conference on Spoken Language Processing

September 16-20, 2002
Denver, Colorado, USA

A Hybrid HMM/Traps Model for Robust Voice Activity Detection

Brian Kingsbury (1), Pratibha Jain (2), Andre Adami (2)

(1) IBM T.J. Watson Research Center, USA; (2) Oregon Health & Science University, USA

We present three voice activity detection (VAD) algorithms that are suitable for the off-line processing of noisy speech and compare their performance on SPINE-2 evaluation data using speech recognition error rate as the quality metric. One VAD system is a simple HMM- based segmenter that uses normalized log-energy and a degree of voicing measure as raw features. The other two VAD systems focus on frequency-localized temporal information in the speech signal using a TempoRAl PatternS (TRAPS) classifier. They differ only in the processing of the TRAPS output. One VAD system uses median filtering to generate segment hypotheses, while the other is a hybrid system that uses a Viterbi search identical to that used in the HMM segmenter. Recognition on the hybrid HMM/TRAPS segmentation is more accurate than recognition on the other two segmentations by 1% absolute. This difference is statistically significant at a 99% con- fidence level according to a matched pairs sentence-segment word error test.


Full Paper

Bibliographic reference.  Kingsbury, Brian / Jain, Pratibha / Adami, Andre (2002): "A hybrid HMM/traps model for robust voice activity detection", In ICSLP-2002, 1073-1076.