Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

On the application of reverberation suppression to robust speech recognition

: Maas, R.; Habets, E.A.P.; Sehr, A.; Kellermann, W.


IEEE Signal Processing Society:
IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2012. Vol.1 : Kyoto, Japan, 25 - 30 March 2012; Proceedings
New York, NY: IEEE, 2012
ISBN: 978-1-4673-0046-9
ISBN: 978-1-4673-0045-2 (Print)
ISBN: 978-1-4673-0044-5 (Online)
International Conference on Acoustics, Speech and Signal Processing (ICASSP) <37, 2012, Kyoto>
Conference Paper
Fraunhofer IIS ()

In this paper, we study the effect of the design parameters of a single-channel reverberation suppression algorithm on reverberation-robust speech recognition. At the same time, reverberation compensation at the speech recognizer is investigated. The analysis reveals that it is highly beneficial to attenuate only the reverberation tail after approximately 50 ms while coping with the early reflections and residual late-reverberation by training the recognizer on moderately reverberant data. It will be shown that the overall system at its optimum configuration yields a very promising recognition performance even in strongly reverberant environments. Since the reverberation suppression algorithm is evidenced to significantly reduce the dependency on the training data, it allows for a very efficient training of acoustic models that are suitable for a wide range of reverberation conditions. Finally, experiments with an ideal reverberation suppression algorithm are carried out to cross-check the inferred guidelines.