First Call for Papers 

Third International Workshop on EMOTION (satellite of LREC):

CORPORA FOR RESEARCH ON EMOTION AND AFFECT 

http://emotion-research.net/sigs/speech-sig/emotion-workshop 

Sunday, 23rd May 2010

Mediterranean Conference Centre- Valletta Malta



In Association with

7th INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION

LREC2010 http://www.lrec-conf.org/lrec2010/

Main Conference 19th-21st May 2010




 

Recognition of emotion in speech has recently matured to one of the key disciplines in speech analysis serving next generation human-machine and ?robot communication and media retrieval systems. 
However, compared to automatic speech and speaker recognition, where several hours of speech of a multitude of speakers in a great variety of different languages are available, sparseness of resources 
has accompanied emotion research to the present day: genuine emotion is hard to collect, ambiguous to annotate, and tricky to distribute due to privacy preservation.

The few available corpora suffer from a number of issues owing to the peculiarity of this young field: as in no related task, different forms of modelling reaching from discrete over complex to continuous 
emotions exist, and ground truth is never solid due to the often highly different perception of the mostly very few annotators. Given by the data sparseness ? most widely used corpora feature below 30 min 
of speech ? cross-validation without strict test, development, and train partitions, and without strict separation of speakers throughout partitioning are the predominant evaluation strategy, which is obviously 
sub-optimal. Acting of emotions was often seen as a solution to the desperate need for data, which often resulted in further restrictions such as little variation of spoken content or few speakers. As a result, 
many interesting potentially progressing ideas cannot be addressed, as clustering of speakers or the influence of languages, cultures, speaker health state, etc..

Previous LREC workshops on Corpora for research on Emotion and Affect (at LREC 2006 and 2008) have helped to consolidate the field, and in particular there is now growing experience of not only 
building databases but also using them to build systems (for both synthesis and detection). This workshop aims to continue the process, and lays particular emphasis on showing how databases can be or
 have been used for system building. 

Papers are invited in the area of corpora for research on emotion and affect. Topics include, but are not limited to: 

+  Novel corpora of affective speech in audio and multimodal data ? in particular with high number of speakers and high diversity (language, age, speaking style, health state, etc.)

+  Case studies of the way databases have been or can be used for system building

+  Measures for quantitative corpus quality assessment 

+  Standardisation of corpora and labels for cross-corpus experimentation

+  Mixture of emotions (i.e. complex or blended emotions)

+  Real-life applications

+  Long-term recordings for intra-speaker variation assessment

+  Rich and novel annotations and annotation types

+  Communications on testing protocols

+  Evaluations on novel or multiple corpora


ORGANISING COMMITEE

Laurence Devillers / Björn Schuller LIMSI-CNRS, France

Roddy Cowie / Ellen Douglas-Cowie Queen's University, UK

Anton Batliner Universität Erlangen-Nürnberg, Germany

Contact: Laurence Devillers and Björn Schuller, lrec-emotion@limsi.fr 


IMPORTANT DATES

 Deadline for 1500-2000 words abstract submission    12th  February

Notification of acceptance                                            12th  March

Final version of accepted paper                                    22nd March

Workshop full-day                                                       23rd  May




SUBMISSIONS

 The workshop will consist of paper and poster presentations. Submitted abstracts of papers for oral and poster must consist of about 1500-2000 words. Final submissions should be 4 pages long, 
must be in English, and follow the submission guidelines at LREC2010. Papers need to be submitted via the START page of LREC 2010. When submitting a paper from the START page, authors 
will be asked to provide essential information about resources (in a broad sense, i.e. also technologies, standards, evaluation kits, etc.) that have been used for the work described in the paper or are 
a new result of your research. For further information on this new iniative, please refer to http://www.lrec-conf.org/lrec2010/?LREC2010-Map-of-Language-Resources

Following this initiative, all contributions shall provide an additional corpus description according to a template (with example) provided by the organisers at the time of submission. The information will 
consist of providing site, domain, classes or dimensions with definition, context, language(s), spoken content, type, status, size, speaker and instance numbers, total duration, recording, 
encoding and storage details, annotator number, annotation state and format, and partitioning type. 

In addition they are asked to provide audio examples if possible.

As soon as possible, authors are encouraged to send to lrec-emotion@limsi.fr a brief email indicating their intention to participate, including their contact information and the topic they intend to address
in their submissions. Submission site: https://www.softconf.com/lrec2010/EMOTION2010/

Proceedings of the workshop will be printed by the LREC Local Organising Committee. Submitted papers will undergo peer-review.

TIME SCHEDULE AND REGISTRATION FEE


The workshop will consist of a full-day session, and there will be time for collective discussions.

For this full-day Workshop, the registration fee will be specified on  http://emotion-research.net/sigs/speech-sig/emotion-workshop