Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

On the Decorrelation of Filter-Bank Energies in Speech Recognition

Climent Nadeu, Javier Hernando, Monica Gorricho

Universitat Politecnica de Catalunya, Barcelona, Spain

Cepstral coefficients are widely used in speech recognition. In this paper, we claim that they are not the best way of representing the spectral envelope, at least for some usual speech recognition systems. In fact, cepstrum has several disadvantages: poor physical meaning, need of transformation, and low capacity of adaptation to some recognition systems. In this paper, we propose a new representation that significantly outperforms both mel-cepstrum and LPC-cepstrum techniques in both recognition rate and computational cost. It consists of filtering the frequency sequence of filter-bank energies with an extremely simple filter that equalizes the variance of the cepstral coefficients. Excellent results of the new technique using a continuous observation density HMM recognition system and two very different recognition tasks, connected digits and phone recognition, are presented.

Full Paper

Bibliographic reference.  Nadeu, Climent / Hernando, Javier / Gorricho, Monica (1995): "On the decorrelation of filter-bank energies in speech recognition", In EUROSPEECH-1995, 1381-1384.