8th European Conference on Speech Communication and Technology

Geneva, Switzerland
September 1-4, 2003


Entropy-Optimized Channel Error Mitigation with Application to Speech Recognition Over Wireless

Victoria Sanchez, Antonio M. Peinado, Angel M. Gomez, Jose L. Perez-Cordoba

Universidad de Granada, Spain

In this paper we propose an entropy-optimized channel error mitigation technique with a low computational complexity and moderate memory requirements, suitable for transmissions over wireless channels. We apply it to Distributed Speech Recognition (DSR), getting an improvement of around 3% in word accuracy over the recognition performance obtained by the mitigation technique proposed in the ETSI standard for DSR (ETSIES- 201-108 v1.1.2) for bad channel conditions (GSMEP3 error pattern).

Full Paper

Bibliographic reference.  Sanchez, Victoria / Peinado, Antonio M. / Gomez, Angel M. / Perez-Cordoba, Jose L. (2003): "Entropy-optimized channel error mitigation with application to speech recognition over wireless", In EUROSPEECH-2003, 1089-1092.