Combined-order hidden Markov models for reverberation-robust speech recognition

Maas R, Kotha SR, Sehr A, Kellermann W (2012)


Publication Language: English

Publication Status: Published

Publication Type: Conference contribution, Conference Contribution

Publication year: 2012

Pages Range: 167-171

Article Number: 6232918

Event location: Baiona ES

ISBN: 978-1-4673-1878-5

DOI: 10.1109/CIP.2012.6232918

Abstract

In this contribution, the concept of combined-order hidden Markov models (CO-HMMs) is introduced by joining the first-order Markov and the second-order conditional independence assumption. The proposed approach is motivated and evaluated in the context of reverberation-robust automatic speech recognition. Two predecessor-dependent output probability density functions per hidden Markov model (HMM) state are employed in order to explicitly cope with the high inter-frame correlation in presence of reverberation. At the same time, the state duration modeling related to the first-order Markov assumption is addressed by a recently published training procedure based on hard alignment having the significant advantage that any conventional HMM can be efficiently updated to a CO-HMM. The experimental results show a reduction in average entropy as well as in word error rate in reverberant environments compared to conventional HMMs. © 2012 IEEE.

Authors with CRIS profile

How to cite

APA:

Maas, R., Kotha, S.R., Sehr, A., & Kellermann, W. (2012). Combined-order hidden Markov models for reverberation-robust speech recognition. In Proceedings of the 3rd International Workshop on Cognitive Information Processing (CIP) (pp. 167-171). Baiona, ES.

MLA:

Maas, Roland, et al. "Combined-order hidden Markov models for reverberation-robust speech recognition." Proceedings of the 3rd International Workshop on Cognitive Information Processing (CIP), Baiona 2012. 167-171.

BibTeX: Download