Microphone-Array Speech Recognition Via Incremental Map Training (1996)
| Venue: | In Proc. of IEEE ICASSP, volume II |
| Citations: | 9 - 2 self |
BibTeX
@INPROCEEDINGS{Adcock96microphone-arrayspeech,
author = {John Adcock and Yoshihiko Gotoh and Daniel J. Mashao and Harvey F. Silverman},
title = {Microphone-Array Speech Recognition Via Incremental Map Training},
booktitle = {In Proc. of IEEE ICASSP, volume II},
year = {1996},
pages = {897--900}
}
Years of Citing Articles
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
For a hidden Markov model (HMM) based speech recognition system it is desirable to combine enhancement of the acoustical signal and statistical representation of model parameters, ensuring both a high quality speech signal and an appropriately trained HMM. In this paper the incremental variant of maximum a posteriori (MAP) estimation is used to adjust the parameters of a talker-independent HMM-based speech recognition system to accurately recognize speech data acquired with a microphone-array. The approach is novel for a microphone-array speech recognition task in that a robust talker-independent model is derived from a baseline system using a relatively small amount of data for training. The results show that (1) MAP training significantly improves recognition performance compared to the baseline, and (2) beamforming signal enhancement outperforms single-channel enhancement before and after the adaptive MAP training. 1. INTRODUCTION Microphone-array systems have been an area of activ...
