A new concept for feature-domain dereverberation for robust distant-talking ASR

Sehr A, Kellermann W (2007)


Publication Language: English

Publication Status: Published

Publication Type: Conference contribution, Conference Contribution

Publication year: 2007

Book Volume: 4

Pages Range: IV-369-IV-372

Article Number: 4218114

Event location: Honolulu, HI US

ISBN: 1-4244-0727-3

DOI: 10.1109/ICASSP.2007.366926

Abstract

The feature-domain dereverberation capabilities of a novel approach for automatic speech recognition in reverberant environments are investigated in this paper. By combining a network of clean speech HMMs and a reverberation model, the most likely combination of the HMM output and the reverberation model output is found during decoding time by an extended version of the Viterbi algorithm. We show in this paper that the most likely HMM output represents a good estimate of the clean speech feature sequence and can be used as input to subsequent speech recognizers. © 2007 IEEE.

Authors with CRIS profile

How to cite

APA:

Sehr, A., & Kellermann, W. (2007). A new concept for feature-domain dereverberation for robust distant-talking ASR. In Proceedings of the 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07 (pp. IV-369-IV-372). Honolulu, HI, US.

MLA:

Sehr, Armin, and Walter Kellermann. "A new concept for feature-domain dereverberation for robust distant-talking ASR." Proceedings of the 2007 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '07, Honolulu, HI 2007. IV-369-IV-372.

BibTeX: Download