ACM Multimedia 2010 SSCS Workshop

The 2010 Workshop on Searching Spontaneous Conversational Speech is devoted to presentation and discussion of recent research results concerning advances and innovation in the area of spoken content retrieval and the area of multimedia search that makes use of automatic speech recognition technology.

The workshop is co-located with ACM Multimedia 2010 and will take place on Friday 29 October, 2010 in Firenze, Italy. For more information on the venue and the exact location of the workshop, please see the ACM Multimedia 2010 Workshop Page.

SSCS 2010 Workshop Program

8:00 Registration
8:30 Welcome
8:45 Searching Speech Real-World Needs (Part I)
Invited Talk: Speech Retrieval for Interview Data: Technology Push versus Academic Demand, Stef Scagliola, (Netherlands Veterans Institute)
9:30 Boasters for demonstration presentations
Three-minute presentations introducing each demo+poster presentation.
  • The ACLD: Speech-based Just-in-Time Retrieval of Meeting Transcripts, Documents and Websites, Andrei Popescu-Belis (Idiap Research Institute, Switzerland), Jonathan Kilgour (University of Edinburgh, UK), Alexandre Nanchen (Idiap) and Peter Poller (DFKI, Germany)
  • The Ambient Spotlight: Queryless Desktop Search From Meeting Speech, Jonathan Kilgour, Jean Carletta and Steve Renals (University of Edinburgh, UK)
  • Large Multimedia Archive for World Languages, Peter Wittenburg (MPI for Psycholinguistics, Netherlands)
  • Spoken News Queries over the World Wide Web, Sebastian Stüker, Michael Heck, Katja Renner and Alex Waibel (Karlsruhe Institute of Technology, Germany)
  • A Parallel Meeting Diarist, Gerald Friedland (International Computer Science Institute, USA), Jike Chong (University of California, Berkeley, USA), and Adam Janin (International Computer Science Institute, USA)
10:00 Challenges of Segmentation
Story Segmentation for Speech Transcripts in Sparse Data Conditions, Laurens van der Werff (University of Twente, Netherlands)
10:30 Coffee break
11:00 Challenges of Spontaneous Content
  • Impact of Spontaneous Speech Features on Business Concept Detection: a Study of Call-Centre Data, Charlotte Danesi and Chloé Clavel (EDF R&D, France) 
  • Automatic Indexing of Speech Segments with Spontaneity Levels on Large Audio Database, Richard Dufour, Yannick Esteve and Paul Deleglise (Université du Maine Le Mans, France)
  • Speaker Role Recognition to help Spontaneous Conversational Speech Detection, Benjamin Bigot, Julien Pinquier, Régine André-Obrecht and Isabelle Ferrané (Paul Sabatier University, France)
12:30 Searching Speech Real-World Needs (Part II)
Invited Talk: Searching Large Archives, Roeland Ordelman (Netherlands Institute for Sound and Vision)

13:00 Lunch

14:30 Searching Speech Real-World Needs (Part III)
Invited Talk: World Wide Telecom Web Search, Nitendra Rajput (IBM Research, India)
15:15 Demonstration session
Presentation of the demonstrations that were "boasted" in the morning session. Participants will move between demonstration stations for discussion and interaction. Coffee is planned.
16:15 Challenge of Search
  • Novel Methods for Query Selection and Query Combination in Query-By-Example Spoken Term Detection, Javier Tejedor (HCTLab - Universidad Autonoma de Madrid, Spain), Igor Szoke and Michal Fapso (Speech@FIT, Czech Republic)
  • Direct Posterior Confidence for Out-of-Vocabulary Spoken Term Detection, Dong Wang (EURECOM, France), Simon King (University of Edinburgh, UK), Nicholas W. D. Evans (EURECOM), Joe Frankel (University of Edinburgh, UK) and Raphael Troncy (EURECOM)
  • Towards Methods for Efficient Access to Spoken Content in the AMI Corpus, Gareth J.F. Jones, Maria Eskevich and Agnes Gyarmati (Dublin City University, Ireland)
17:45 Closing words and announcement of Call for Papers for the upcoming ACM TOIS Special Issue on Searching Speech.

Instructions to presenters

Each oral presentation is allocated a 30 minute slot (20 min. presentation + 10 min. questions). Demonstration presentations consists of three parts (1) A three minute "boaster" in which the presenter explains the demo system and why it is interesting. The presenter should prepare 2-3 slides for the boaster (2) A poster covering the main points of the demo (3) the demo itself. You might consider bringing a set of small speakers unless you have exceptionally powerful laptop speakers. Please contact Martha Larson m.a.larson (at) tudelft.nl if you have questions.

Workshop Organizers

Martha Larson Delft University of Technology (Netherlands)
Roeland Ordelman Uni. of Twente and Sound & Vision (Netherlands)
Florian Metze Carnegie Mellon University (USA)
Franciska de Jong University of Twente (Netherlands)
Wessel Kraaij Radboud University Nijmegen and TNO ICT (Netherlands)

Workshop Advisory Board

Joachim Kohler Fraunhofer IAIS (Germany)
Doug Oard University of Maryland (USA)