Evaluation of the Space Denoising Algorithm on AURORA2

Recently we introduced a new and simple denoising algorithm, called SPACE, that yielded promising preliminary results in noise robust speech recognition. SPACE is essentially based on GMM modeling of clean an noisy speech. In this paper, we evaluate the performance of SPACE on Aurora2 and show that...

Full description

Saved in:
Bibliographic Details
Published in2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings Vol. 1; p. I
Main Authors Cerisara, C., Daoudi, K.
Format Conference Proceeding
LanguageEnglish
Published IEEE 2006
Subjects
Acoustic noise
Automatic speech recognition
Gaussian noise
Hidden Markov models
Noise reduction
Noise robustness
Performance evaluation
Speech recognition
Testing
Working environment noise
Online AccessGet full text

Cover

More Information
Summary:Recently we introduced a new and simple denoising algorithm, called SPACE, that yielded promising preliminary results in noise robust speech recognition. SPACE is essentially based on GMM modeling of clean an noisy speech. In this paper, we evaluate the performance of SPACE on Aurora2 and show that they are globally not satisfactory, essentially because the Gaussian correspondence assumption is not verified. We then propose a new training procedure for the GMMs that achieves a better Gaussian correspondence. We further develop a simple adaptation algorithm to handle unknown environments that preserves the Gaussian correspondence. We evaluate the new denoising algorithm on Aurora2. The results show that it outperforms the multistyle models, sometimes significantly, on the three test sets of Aurora2
ISBN:9781424404698
142440469X
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2006.1660072