5th European Conference on Speech Communication and Technology

Rhodes, Greece
September 22-25, 1997

Equalizing Sub-Band Error Rates in Speaker Recognition

Roland Auckenthaler (1), John S. Mason (2)

(1) Department of Electronics, Technical University Graz, Graz, Austria (2)Department of Electrical & Electronic Engineering, University of Wales, Swansea, UK

Recent work in ASR shows that band splitting, forming multiple paths with recombination at the decision stage, can give recognition accuracy comparable with the conventional full-band approach. One of the many interesting questions with band-splitting relates to the bandwidths of each sub-band, and the use of frequency warping functions such as mel. This paper examines the use of mel and linear frequency scales in the context of band-splitting and speaker recognition. We demonstrate how sub-band error profiles can lead to a new scale, which is between linear and mel, giving both an equalised sub-band error profile and an improved overall recognition accuracy.

Full Paper

Bibliographic reference.  Auckenthaler, Roland / Mason, John S. (1997): "Equalizing sub-band error rates in speaker recognition", In EUROSPEECH-1997, 2303-2306.