W3C Workshop on Emotion Markup Language

This W3C Workshop is taking place right before the Accounting for Social Variables in Human Computer Interaction Workshop linked to the European project SSPNet (Social Signal Processing Network).

Background

The W3C Emotion Markup Language (EmotionML) is a representation of emotions and emotion-related states for use in technology. As the web is becoming ubiquitous, interactive, and multimodal, technology needs to deal increasingly with human factors including emotions. The present draft specification of EmotionML aims to strike a balance between practical applicability and scientific well-foundedness, and we believe the language is important for accessibility as well, e.g., helps people with autism spectrum.

EmotionML is based on the understanding that there is no general agreement in the community regarding the vocabularies that should be used for describing emotions. As a consequence, the user can specify which emotion vocabulary to use in a given document. The specification comes with a selection of "recommended" vocabularies; users can choose among these or define their own, custom emotion vocabularies.

We identified three broad use case types for the language.

1. Manual annotation of material involving emotionality, such as annotation of videos, of speech recordings, of faces, of texts, etc.;
2. Automatic recognition of emotions from sensors, including physiological sensors, speech recordings, facial expressions, etc., as well as from multi-modal combinations of sensors;
3. Generation of emotion-related system responses, which may involve reasoning about the emotional implications of events, emotional prosody in synthetic speech, facial expressions and gestures of embodied agents or robots, the choice of music and colors of lighting in a room, etc.

Workshop Goals and Scope

The workshop is aimed at receiving feedback from the community on the current EmotionML specification, with an emphasis on the following issues:

• Is the current list of recommended vocabularies scientifically sound and defendable? Should descriptions be added, removed, or presented differently?
• Does the specification have sufficient expressive power? Can it represent what people need to represent?
• Is EmotionML easy enough to use, or should the syntax be changed somehow to avoid any confusions etc.?

In order to address these questions, it might be helpful to develop for each of the three above-mentioned application types a concrete, protoypical, use case scenario, on the basis of which the issues could be validated. Participants are encouraged to outline such use cases in their position papers as a basis for a common discussion during the workshop.

Position papers and discussions at the workshop are expected to lead to an understanding whether revisions to the EmotionML specification are needed before the formal standardisation process continues.

Requirements for Participation

Position papers will be the basis for the discussions at the workshop. Position papers must be submitted by the date shown below. The program committee will select papers that provide targeted contributions for improving the Emotion Markup Language.

All workshop attendees must submit a position paper of 1 to 5 pages. All position papers will be published on the workshop website; the authors of selected position papers will be invited to present their position paper at the workshop to foster discussion.

Participation in the workshop is conditional upon acceptance of the position paper by the program committee.

Position papers must be written in English. All papers should be 1 to 5 pages, although they may link to longer versions or appendices. Allowed formats are valid HTML or XHTML, PDF, or plain text. Papers in any other format (including invalid HTML/XHTML) will be returned with a request for correct formatting.

Accepted position papers will be published on the public Web page of the workshop. Submitting a position paper comprises a default recognition of these terms for publication.

The Program Committee will ask the authors of particularly salient position papers to explicitly present their position at the workshop to foster discussion. Presenters will be asked to make the slides of the presentation available on the workshop home page in HTML, PDF, or plain text.

See important dates below for submission and registration deadlines.

Participation will be governed by the following:

• To ensure maximum diversity, the number of participants per organization will be limited.
• W3C membership is not required to participate in this workshop.
• Workshop sessions and documents will be in English.
• Attendees are required to submit a position paper.

Workshop Organization

Workshop Organizing Committee:

• Marc Schröder, Editor of EmotionML in the W3C Multimodal Interaction Working Group (DFKI), <marc.schroeder@dfki.de>
• Catherine Pelachaud, Member of the W3C Multimodal Interaction Working Group (CNRS, Telecom ParisTech), <catherine.pelachaud@telecom-paristech.fr>
• Deborah Dahl, Chair of the W3C Multimodal Interaction Working Group (W3C Invited Expert), <dahl@conversational-technologies.com>
• Kazuyuki Ashimura, Multimodal Interaction/Voice Browser Activity Lead (W3C), <ashimura@w3.org>

Program Committee:

The current program committee consists of:

• Paolo Baggia, Loquendo
• Felix Burkhardt, Deutsche Telekom
• Rafael Calvo, NICTA
• Sazzad Hussain, NICTA
• Alessandro Oltramari, CNR
• Catherine Pelachaud, CNRS, Telecom ParisTech
• Christian Peter, Fraunhofer Gesellschaft
• Marc Schröder, DFKI
• Enrico Zovato, Loquendo

Venue and Schedule:

The workshop will be held at Institut Telecom Paristech, Paris, France.

The workshop program will run from 9:00 am to 6 pm on both days, 5 and 6 October 2010.

Registration:

Information on registration, hotel and logistics will be sent with the notification of acceptance.

Important Dates

The Logistics, the Presentation Guideline and the Agenda will be made available on the W3C Web server.