Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

An Elitist Approach for Extracting Automatically Well-Realized Speech Sounds with High Confidence

Jean-Baptiste Maj, Anne Bonneau, Dominique Fohr, Yves Laprie

LORIA, Vandoeuvre-les-Nancy, France

This paper presents an 'elitist approach' for extracting automatically well-realized speech sounds with high confidence. The elitist approach uses a speech recognition system based on Hidden Markov Models (HMM). The HMM are trained on speech sounds which are systematically well-detected in an iterative procedure. The results show that, by using the HMM models defined in the training phase, the speech recognizer detects reliably specific speech sounds with a small rate of errors.

Full Paper

Bibliographic reference.  Maj, Jean-Baptiste / Bonneau, Anne / Fohr, Dominique / Laprie, Yves (2005): "An elitist approach for extracting automatically well-realized speech sounds with high confidence", In INTERSPEECH-2005, 2925-2928.