4th International Conference on Spoken Language Processing Philadelphia, PA, USA October 3-6, 1996

Robust Gender-dependent Acoustic-phonetic Modelling in Continuous Speech Recognition Based on a New Automatic Male/Female Classification

Rivarol Vergin, Azarshid Farhat, Douglas O'Shaughnessy

INRS-Telecommunications, Ile-des-Surs, Québec, Canada

In this paper we present a new automatic male/female classification method based on the location in the frequency domain of the first 2 formants. This classification is based on a new automatic formant extraction which is faster than a peak picking technique. Gender-dependent acoustic-phonetic models stemming from this classification are used in the INRS Continuous speech recognition system with ATIS corpora. An improvement of 14% is obtained with these models in comparison to the baseline speaker-independent system.

Full Paper

Bibliographic reference.  Vergin, Rivarol / Farhat, Azarshid / O'Shaughnessy, Douglas (1996): "Robust gender-dependent acoustic-phonetic modelling in continuous speech recognition based on a new automatic male/female classification", In ICSLP-1996, 1081-1084.