Probabilistic vector mapping of noisy speech parameters for HMM word spotting
- 4 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 117-120
- https://doi.org/10.1109/icassp.1990.115552
Abstract
A conditional probability model is developed for relating a noisy, observation feature vector to the noise-free vector that generated it. The model is a Gaussian mixture which is based on the vectors and is conditioned on the instantaneous signal-to-noise ratio at the frame. When the feature vector estimates based on this model are used in a hidden Markov model (HMM) word spotter trained with noise-free speech, a performance gain of about 20%-30% is observed (depending on spotter topology) compared to that of the HMM word spotter trained with noisy speech.Keywords
This publication has 3 references indexed in Scilit:
- Continuous hidden Markov modeling for speaker-independent word spottingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Spectral estimation for noise robust speech recognitionPublished by Association for Computational Linguistics (ACL) ,1989
- Speech recognition using noise-adaptive prototypesIEEE Transactions on Acoustics, Speech, and Signal Processing, 1989