A highly efficient optimization scheme for remos-based distant-talking speech recognition

Conference contribution
(Conference Contribution)


Publication Details

Author(s): Maas R, Sehr A, Gugat M, Kellermann W
Publication year: 2010
Pages range: 1983-1987
ISSN: 2219-5491
Language: English


Abstract


A highly efficient decoding algorithm for the REMOS (REverberationMOdeling for Speech recognition) concept for distant-talking speech recognition as proposed in [1] is suggested to reduce the computational complexity by about two orders of magnitude and thereby allowing for first real-time implementations. REMOS is based on a combined acoustic model consisting of a conventional hidden Markov model (HMM), modeling the clean speech, and a reverberation model. During recognition, the most likely clean-speech and reverberant contributions are estimated by solving an inner optimization problem for logarithmic melspectral (logmelspec) features. In this paper, two approximation techniques for the inner optimization problem are derived. Connected digit recognition experiments confirm that the computational complexity is significantly reduced. Ensuring that the global optima of the inner optimization problem are found, the decoding algorithm based on the proposed approximations even increases the recognition accuracy relative to interior point optimization techniques. © EURASIP, 2010.


FAU Authors / FAU Editors

Gugat, Martin apl. Prof. Dr.
Lehrstuhl für Angewandte Mathematik
Kellermann, Walter Prof. Dr.-Ing.
Professur für Nachrichtentechnik
Maas, Roland
Lehrstuhl für Multimediakommunikation und Signalverarbeitung
Sehr, Armin Dr.-Ing.
Professur für Nachrichtentechnik


How to cite

APA:
Maas, R., Sehr, A., Gugat, M., & Kellermann, W. (2010). A highly efficient optimization scheme for remos-based distant-talking speech recognition. In Proceedings of the 18th European Signal Processing Conference, EUSIPCO 2010 (pp. 1983-1987). Aalborg, DK.

MLA:
Maas, Roland, et al. "A highly efficient optimization scheme for remos-based distant-talking speech recognition." Proceedings of the 18th European Signal Processing Conference, EUSIPCO 2010, Aalborg 2010. 1983-1987.

BibTeX: 

Last updated on 2019-29-04 at 21:50