De la reconnaissance automatique de la parole à l ’ analyse linguistique de corpus oraux

This contribution aims at giving an overview of present automatic speech recognition in French highlighting typical transcription problems for this language. Explanations for errors can be partially obtained by examining the acoustics of the speech data. Such investigations however do not only inform about system and/or speech modeling limitations, but they also contribute to discover, describe and quantify specificities of spoken language as opposed to written language and speakers’ speech performance. To automatically transcribe different speech genres (e.g. broadcast news vsconversations) specific acoustic corpora are used for training, suggesting that frequency of occurrence and acoustic realisations of phonemes vary significantly across genres. Some examples of corpus studies are presented describing phoneme frequencies and segment durations on different corpora. In the future largescale corpus studies may contribute to increase our knowledge of spoken language as well as the performance of automatic processing.

[1]  Lori Lamel,et al.  Investigating syllabic structures and their variation in spontaneous French , 2005, Speech Commun..

[2]  Alexandre Allauzen,et al.  Where are we in transcribing French broadcast news? , 2005, INTERSPEECH.

[3]  Mark Liberman,et al.  Transcriber: Development and use of a tool for assisting speech corpora production , 2001, Speech Commun..

[4]  Maxine Eskénazi,et al.  BREF, a large vocabulary spoken corpus for French , 1991, EUROSPEECH.

[5]  Chantal Lyche,et al.  Le projet "Phonologie du français contemporain" (PFC) , 2003 .

[6]  Lori Lamel,et al.  PRONUNCIATION VARIANTS IN FRENCH : SCHWA & LIAISON , 1999 .

[7]  P. Fouché.  Traité de prononciation française , 1956 .

[8]  Lori Lamel,et al.  Speaker-independent continuous speech dictation , 1993, Speech Communication.

[9]  Pierre Encrevé,et al.  La liaison avec et sans enchaînement : phonologie tridimensionnelle et usages du français , 1988 .

[10]  Lori Lamel,et al.  Issues in Large Vocabulary, Multilingual Speech Recognition , 1995, EUROSPEECH.

[11]  F. Dell,et al.  Les règles et les sons : introduction à la phonologie générative , 1985 .

[12]  Danielle Duez.  MODELLING ASPECTS OF REDUCTION AND ASSIMILATION OF CONSONANT SEQUENCES IN SPONTANEOUS FRENCH SPEECH , 2007 .

[13]  André Malécot.  Frequency of Occurrence of French Phonemes and Consonant Clusters , 1974 .

[14]  Guillaume Gravier,et al.  The ESTER phase II evaluation campaign for the rich transcription of French broadcast news , 2005, INTERSPEECH.

[15]  Lori Lamel,et al.  Do speech recognizers prefer female speakers? , 2005, INTERSPEECH.

[16]  Patrick Paroubek,et al.  A quantitative study of disfluencies in French broadcast interviews , 2005, DiSS.

[17]  Martine Adda-Decker,et al.  Liaisons in French: a corpus-based study using morpho-syntactic information , 2003 .

[18]  Lori Lamel,et al.  Pronunciation variants across system configuration, language and speaking style , 1999, Speech Commun..

[19]  Cédric Gendrot,et al.  Impact of duration on F1/F2 formant values of oral vowels: an automatic analysis of large broadcast news corpora in French and German , 2005, INTERSPEECH.