International workshop on audio-visual affective prosody in social interaction & second language learning (AVAP 2015)


March 5th and 6th 2015

This workshop focuses on social affective meaning (including emotions, mood, feeling and intentions or attitudes) conveyed by audio as well as visual control. Expressiveness of affective meanings in many languages / cultures involves research in various scientific fields: affective computing, cultural psychology, cognitive psychology, linguistic, phonetics, speech processing, visual processing and artificial intelligence. This workshop will bring together specialists of these fields.

More information

International Conference: Context-based Spoken Japanese Language


April 4th and 5th 2014

It is well known that a speaker adapts his speech according to the context and circumstances. However, linguists have few concepts, other than talking about "style", for conducting analyses of how subjects adapt their form of discourse to speech communication goals; it is clear that a traditional approach cannot cover the multiplicity and plurality of the phenomena involved.

In recent years, a new perspective has been opened in Japan with regard to this: the development and delineation of an original research field around the concept of "kyarakutaa" (or "kyara" in its abbreviated form of English terminology "character"). The term "kyarakutaa / character / character" is not limited to linguistic phenomena in a narrow, traditional sense; it affects phonetic dimensions, gestures, behaviour and even the cultural whole.

How does the notion of "kyarakutaa / character / personage" allow us to see the issues of communication in a new light? How does this notion explain in a more operational way than the way given by the notion of "style" to understand the diversity of language variations incurred by the speaker to adapt to context and circumstances?

The international conference to be held in Bordeaux April 4th and 5th, 2014 will discuss these issues, centering around the work by Professor Toshiyuki Sadanobu, Kobe University, Japan. Professor Sadanobu is a pioneer and leading expert in the research field of discourse in context and specifically, "kyarakutaa (character)". He will be accompanied by several members of his research team who will also present their ongoing research.

The meetings in Bordeaux are expected to provide the occasion to meet researchers involved in the fields of linguistics and Japanese culture, both inside and outside of Japan. We are looking forward to receiving proposals to present about "Context-based spoken Japanese language" in all its aspects: lexical, syntactic, phonetic, socio-linguistic, stylistic as well as other aspects in Japanese linguistics or in Japanese teaching methods.

More information

Spectacle "Fluxus"


Représentations : August 4, 2013, March 16 2013, March 16 2012

Fluxus - île de Patiras - 4 août 2013

Dans le cadre d'Histoires d'îles, les Tafurs et le SCRIME présentaient à nouveau le spectacle Fluxus, cette fois-ci en extérieur sur l'île de Patiras. Jeanne Flora et François Mauget des Tafurs accueillaient les spectateurs arrivant par bateau avec des textes de marins contés pendant la traversée.

« Fluxus » raconte l'histoire métaphorique d'un fleuve, depuis sa source jusqu'à la mer. Les spectateurs sont immergés dans un univers onirique mêlant la voix des acteurs aux machines sonores du compositeur Gyorgy Kurtag. Une invitation à naviguer entre fiction poétique et présence évocatrice du fleuve.

Video excerpt

International Workshop of Cross-Cultural research on Speech Communication and Second Language learning processing


Friday, March 15th 2013

I am co-organising this workshop with Takaaki Shochi, Makoto Asari, Laurence Labrune, Catherine Mathon and Albert Rilliard.

More information



Jean-Luc Rouas obtained his PhD degree from Université Paul Sabatier in Toulouse, France, in 2005. He then worked as a research associate at the French National Institute for Transports and Safety Research (INRETS) in Villeneuve d'Ascq (Northern France) in 2005-2006 and 2007-2008, and at the Spoken Language Systems Lab (L2F) in Lisbon, Portugal (2006). He has been a CNRS Researcher since 2008. He first joined the LIMSI (Orsay) and now is at LaBRI (Bordeaux).

His PhD thesis is about automatic language identification using prosodic cues. His work interests range from audio events detection (as for security purposes during one of his post-doctoral positions) to speaking styles characterisation, singing voice and musical information retrieval.



Après l'obtention de son doctorat en informatique de l’université Paul Sabatier (Toulouse 3) en mars 2005, Jean-Luc Rouas a été post-doctorant à l’Institut de Recherche sur les Transports et leur Sécurité à Villeneuve d’Ascq (2005-2006 et 2007-2008) ainsi qu'au Laboratoire de Traitement de la Parole (L2F) à Lisbonne (2006). Il est Chargé de Recherches au CNRS depuis le 01/10/2008, affecté de 2008 à 2010 au LIMSI (Orsay) et depuis 2010 au LaBRI.

Jean-Luc Rouas a effectué sa thèse sur la problématique de l’identification automatique des langues en utilisant des indices prosodiques. Il a depuis effectué des séjours post-doctoraux à l’Institut de Recherche sur les Transports et leur Sécurité (INRETS) au sein du Laboratoire Electronique, Ondes et Signaux pour les Transports (LEOST), cadre dans lequel il a travaillé sur la détection de sons clés dans les environnements de transport, et au Laboratoire de Traitement de la Parole (L2F) à Lisbonne, Portugal, où il a continué ses travaux sur l’identification des langues et des dialectes. Au LIMSI, il a également travaillé sur la caractérisation des styles de paroles. Depuis son affectation au Labri, il s'intéresse au traitement automatique de la voix chantée ainsi qu'à la recherche d'information musicales.