Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

Time Derivatives, Cepstrai Normaiization, and Spectral Parameter Filtering for Continuously Spelled Names over the Telephone

Jean-Claude Junqua (1), Dominique Fohr (2), J.-F. Mari (2), Ted H. Applebaum (1), Brian A. Hanson (1)

(1) Speech Technology Laboratory, Panasonic Technologies Inc., USA
(2) CRIN-CNRS & INRIA Lorraine, France

In this paper, we focus on spectral parameter filtering for reducing the mismatch between training and testing and report experimental results on a continuously spelled name recognition task over the telephone. We studied various time derivative feature combinations, the influence of RASTA processing, short-term and long-term cepstrai mean normalization, and the influence of the amount of training data on recognition performance. Based on the results of these experiments, we derived a new front-end for our task, leading to an error rate improvement of almost 30% in name retrieval as compared to previous published results. We also discuss the interaction between the different techniques studied when used in combination.

Full Paper

Bibliographic reference.  Junqua, Jean-Claude / Fohr, Dominique / Mari, J.-F. / Applebaum, Ted H. / Hanson, Brian A. (1995): "Time derivatives, cepstrai normaiization, and spectral parameter filtering for continuously spelled names over the telephone", In EUROSPEECH-1995, 1385-1388.