Robustesse et portabilités multilingue et multi-domaines des systèmes de compréhension de la parole : les corpus du projet PortMedia (Robustness and portability of spoken language understanding systems among languages and domains : the PORTMEDIA project) [in French]

Le projet ANR PORTMEDIA avait pour objectif de completer le corpus MEDIA afin de favoriser le developpement de methodes performantes, notamment statistiques, pour la comprehension automatique de la parole dans le cadre des systemes de dialogues homme-machines. Les principaux axes traites sont : la robustesse aux erreurs de reconnaissance de la parole, la portabilite multilingue, la portabilite multi-domaines et la representation semantique haut-niveau. Ainsi tout en elaborant au sein du projet des elements de solution a ces problematiques nous nous sommes principalement attaches a elaborer des donnees et meta-donnees permettant ensuite a d’autres groupes de recherche d’evaluer dans les meilleures conditions possibles leurs propres propositions.

[1]  Frédéric Béchet,et al.  The French MEDIA/EVALDA Project: the Evaluation of the Understanding Capability of Spoken Language Dialogue Systems , 2004, LREC.

[2]  Sophie Rosset,et al.  Semantic annotation of the French media dialog corpus , 2005, INTERSPEECH.

[3]  Thierry Bazillon,et al.  Using MMIL for the High Level Semantic Annotation of the French MEDIA Dialogue Corpus , 2011, IWCS.

[4]  Giuseppe Riccardi,et al.  How may I help you? , 1997, Speech Commun..

[5]  Paul Deléglise,et al.  Improvements to the LIUM French ASR system based on CMU sphinx: what helps to significantly reduce the word error rate? , 2009, INTERSPEECH.

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

[7]  Matthieu Quignard,et al.  An Incremental Architecture for the Semantic Annotation of Dialogue Corpora with High-Level Structures. A case of study for the MEDIA corpus , 2011, SIGDIAL Conference.

[8]  Jean-Luc Gauvain,et al.  Genericity and portability for task-independent speech recognition , 2005, Comput. Speech Lang..

[9]  Frédéric Béchet,et al.  The EPAC Corpus: Manual and Automatic Annotations of Conversational Speech in French Broadcast News , 2010, LREC.

[10]  Fabrice Lefèvre,et al.  Investigating multiple approaches for SLU portability to a new language , 2010, INTERSPEECH.

[11]  Fabrice Lefèvre,et al.  Cross-lingual spoken language understanding from unaligned data using discriminative classification models and machine translation , 2010, INTERSPEECH.

[12]  Guillaume Gravier,et al.  The ester 2 evaluation campaign for the rich transcription of French radio broadcasts , 2009, INTERSPEECH.