Fourth European Conference on Speech Communication and Technology

Madrid, Spain
September 18-21, 1995

Speaker-Adaptation for Hybrid HMM-ANN Continuous Speech Recognition System

Joao Neto (1), Luis Almeida (1), Mike Hochberg (2), Ciro Martins (1), Luis Nunes (1), Steve Renals (2), Tony Robinson (2)

(1) Instituto de Engenharia de Sistemas e Computadores (INESC), Portugal, Instituto Superior Tecnico (IST), Portugal
(2) Cambridge University Engineering Department (CUED), UK, Sheffield University (SU), UK

It is well known that recognition performance degrades significantly when moving from a speaker-dependent to a speaker-independent system. Traditional hidden Markov model (HMM) systems have successfully applied speaker-adaptation approaches to reduce this degradation. In this paper we present and evaluate some techniques for speaker-adaptation of a hybrid HMM-artificial neural network (ANN) continuous speech recognition system. These techniques are applied to a well trained, speaker-independent, hybrid HMM-ANN system and the recognizer parameters are adapted to a new speaker through off-line procedures. The techniques are evaluated on the DARPA RM corpus using varying amounts of adaptation material and different ANN architectures. The results show that speaker-adaptation within the hybrid framework can substantially improve system performance.

Full Paper

Bibliographic reference.  Neto, Joao / Almeida, Luis / Hochberg, Mike / Martins, Ciro / Nunes, Luis / Renals, Steve / Robinson, Tony (1995): "Speaker-adaptation for hybrid HMM-ANN continuous speech recognition system", In EUROSPEECH-1995, 2171-2174.