Acoustic-Phonetic Modeling of Non-Native Speech for Language Identification
Wanneroy, R., Éric Bilinski, Clément Barras, Martine Adda‐Decker, Edouard Geoffrois · 1999
The aim of this paper is to investigate to what extent non nativespeechmaydeteriorate language identi cation (LID) performances and to improve them using acoustic adaptation. Our reference LID system is based on a phonotactic approach. The system makes use of language-independent acoustic models and language-speci c phone-based bigram language models. Experiments are conducted on the SQALE test database, which contains recordings from English, French and German nativespeakers, and on the MIST database, which contains non-native speech in the same languages uttered by Dutch speakers. Using 5 seconds of telephone quality speech, language identi cation error rate amounts to 19 % for native speech and to 31 % for non-native speech, thus yielding about 60 % relative error rate increase. Eventually we propose to improve non-native language identi-cation by an adaptation of the acoustic models to the non-native speech. 1