PROSLI - Prosody for self-learning instruction

Workshop at ICNAAM

26-27 September 2013, Rodos Palace Hotel, Rhodes, Greece

                 

Prosody addresses fundamental components of  speech communication - such as intonation and rhythm  - which must be mastered by language learners in order to understand and make themselves understood  properly in the target language. Automatic Speech  Recognition (ASR) almost never incorporates such  components. As a result, speech technologies for  self-learning instructions (SLI) are unable to offer  to learners as to their prosodic skills. In  most cases, prosody is made object of specific  separate tools. 

The workshop will try to cover the whole spectrum of topics both theoretical and practical, that deal  with prosody and self-learning instructions. It will include demonstration and state of the art of such  tools, their effectiveness in self-learning contexts. It will address prosodic issues in   language learning from both the analysis and the perception viewpoint and their accountability in real world applications. In particular, of interest  to the workshop will also be ASR systems for languages like Chinese, which naturally incorporate tone level parameters in their language models; or like Japanese which need to account for durational differences at phonemic and subphonemic level.

                 

Submission: look for formatting details in the main conference website, and then send a 4 pages extended abstract with your affiliation to this  address, delmont@unive.it.

                 

Deadlines

•Early-bird                     registration: April 20, 2013

•Paper                     submission: May 30, 2013

•Notification of                     acceptance: June 30, 2013

•Submission of                     camera ready paper: July 28, 2013

•Conference                     days: September 26-27, 2013

                 

Program Committee

Björn Granström, KTH, Center Speech Technology

Daniel Hirst, CNRS & Aix-Marseille Université

Philippe Martin, Université Paris Diderot

Maxine Eskenazi, CMU, USA

Isabel Trancoso, IST-INESC, Lisbon

Ruediger Hoffmann - Technische Universität                     Dresden

Helmer Strik, Radboud University, Nijmegen

Anton Batliner, University of Munich

Barbara Gili Favela, Università di Lecce

Fabio Tamburini, Università di Bologna

Mariapaola D'Imperio, CNRS & Aix-Marseille                     Université

Jared Bernstein, Pearson Knowledge Technologies,                     USA

Brigitte Bigi, CNRS & Aix-Marseille                     Université

Anne Bonneau, LORIA, France

Uwe Reichel, University of Munich

Rodolfo Delmonte, Ca' Foscari University Venice

                 

                 

Prosody for self-learning instruction

Prosody addresses fundamental components of                     speech communication - such as intonation and rhythm                     - which must be mastered by language learners in                     order to understand and make themselves understood                     properly in the target language. Automatic Speech                     Recognition (ASR) almost never incorporates such                     components. As a result, speech technologies for                     self-learning instructions (SLI) are unable to offer                     feedback to learners as to their prosodic skills. In                     most cases, prosody is made object of specific                     separate tools. 

The workshop will try to cover the whole spectrum                     of topics both theoretical and practical, that deal                     with prosody and self-learning instructions. It will                     include demonstration and state of the art of such                     tools, their effectiveness in self-learning                     contexts. It will address prosodic issues in                     language learning from both the analysis and the                     perception viewpoint and their accountability in                     real world applications. In particular, of interest                     to the workshop will also be ASR systems for                     languages like Chinese, which naturally incorporate                     tone level parameters in their language models; or                     like Japanese which need to account for durational                     differences at phonemic and subphonemic level.

                 

                 

A list of papers will be selected for publication                     of an extended and revised version in a book by                     Springer of the series Studies in Computational                     Intelligence.