Third European Conference on Speech Communication and Technology

Berlin, Germany
September 22-25, 1993


Segmental Post-Processing of the N-Best Solutions in a Speech Recognition System

M. N. Lokbani, D. Jouvet, J. Monne

France Telecom, Centre National d'Etudes des Telecommunications CNET/LAA/TSS/RCP, Lannion, France

This paper presents a new statistical segmental post-processing of the N-best solutions obtained by a hidden Markov model based speech recognizer. This post-processing computes, for each solution, a new segmental score which is latter combined with the markovian score. The solution with the highest score identifies the answer. We introduce a new method in the statistical modelling of each segment involving two models: one for correct segmentation and one for incorrect segmentation. The first results presented here shows that this new approach leads to a 10 to 15 % reduction of the error rate.

Keywords: N-best solutions, segmental post-processing, statistical modelling.

Full Paper

Bibliographic reference.  Lokbani, M. N. / Jouvet, D. / Monne, J. (1993): "Segmental post-processing of the n-best solutions in a speech recognition system", In EUROSPEECH'93, 811-814.