Probabilistic vector mapping of noisy speech parameters for HMM word spotting

Author

Gish, Herbert ; Chow, Yen-Lu ; Rohlicek, J. Robin

Author_Institution

BBN Syst. & Technol. Corp., Cambridge, MA, USA

fYear

1990

fDate

3-6 Apr 1990

Firstpage

117

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

speech analysis and processing; speech recognition; Gaussian mixture; HMM; HMM word spotting; conditional probability model; feature vector estimates; hidden Markov model; instantaneous signal-to-noise ratio; noise-free vector; observation feature vector; performance gain; probabilistic vector mapping of noisy speech parameters; word recognition; word spotter trained with noise-free speech; word spotter trained with noisy speech; Background noise; Cepstral analysis; Hidden Markov models; Noise generators; Noise level; Performance gain; Signal processing; Signal to noise ratio; Space technology; Speech; Speech enhancement; Speech processing; Topology; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on

Conference_Location

Albuquerque, NM

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.1990.115552

Filename

115552