Noise-robust HMMs based on minimum error classification

Authors:
Kazumi Ohkura;David Rainton;Masahide Sugiyama
Affiliations:
ATR Interpreting Telephony Research Laboratories, Kyoto, Japan;ATR Interpreting Telephony Research Laboratories, Kyoto, Japan;ATR Interpreting Telephony Research Laboratories, Kyoto, Japan
Venue:
ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: speech processing - Volume II
Year:
1993

Citing 3
Cited 1

Introduction to non-linear optimization

Introduction to non-linear optimization
Vector equalization in hidden Markov models for noisy speech recognition

ICASSP'92 Proceedings of the 1992 IEEE international conference on Acoustics, speech and signal processing - Volume 1
Segmental GPD training of HMM based speech recognizer

ICASSP'92 Proceedings of the 1992 IEEE international conference on Acoustics, speech and signal processing - Volume 1

Noisy speech recognition performance of discriminative HMMs

ISCSLP'06 Proceedings of the 5th international conference on Chinese Spoken Language Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper compares and contrasts the noise robustness of HMMs trained using a discriminant minimum error classification(MEC) optimization criterion, against that of HMMs trained using the conventional maximum likelihood (ML) approach. Isolated word recognition experiments, performed on the ATR 5240 Japanese word database, gave the following results: 1) MEC continuous Gaussian mixture density HMMs, trained in a specific noisy environment, were more robust to changes in the signal-to-noise (SNR) ratio, than conventional ML HMMs and 2) MEC HMMs, trained in various noisy environments, were more robust in all environments than conventional ML HMMs.