Semi-Automatic Calibration for Dereverberation by Spectral Subtraction for Continuous Speech Recognition

  • In this article, we describe a semi-automatic calibration algorithm for dereverberation by spectral subtraction. We verify the method by a comparison to a manual calibration derived from measured room impulse responses (RIR). We conduct extensive experiments to understand the effect of all involved parameters and to verify values suggested in the literature. The experiments are performed on a text read by 31 speakers and recorded by a headset and three far-field microphones. Results are measured in terms of automatic speech recognition (ASR) performance using a 1-gram model to emphasize acoustic recognition performance. To accommodate for the acoustic change by dereverberation we apply supervised MAP adaptation to the hidden Markov model output probabilities. The combination of dereverberation and adaptation yields a relative improvement of about 35% in terms of word error rate (WER) compared to the original signal.

Export metadata

Additional Services

Search Google Scholar
Author:Korbinian Riedhammer, Tobias Bocklet, Juan Rafael Orozco-Arroyave, Elmar Nöth
Parent Title (English):ITG Symposium on Speech Communication 2014, Erlangen.
Document Type:Conference Proceeding
Publication Year:2014
Tag:Speech Recognition
faculties / departments:Fakultät für Informatik
Dewey Decimal Classification:0 Informatik, Informationswissenschaft, allgemeine Werke / 00 Informatik, Wissen, Systeme / 000 Informatik, Informationswissenschaft, allgemeine Werke