[Subject Prev][Subject Next][Thread Prev][Thread Next][Date Index][Thread Index]

[hts-users:03078] The Listening Talker: an interdisciplinary workshop on natural and synthetic modification of speech in response to listening conditions


The Listening Talker: an interdisciplinary workshop on natural and synthetic modification of speech in response to listening conditions

Edinburgh, 2-3 May 2012

http://listening-talker.org/workshop

When talkers speak, they also listen. Talkers routinely adapt to their interlocutors and environment, maintaining intelligibility and dialogue fluidity in a way that promotes efficient exchange of information. In contrast, current speech output technology is largely deaf, incapable of adapting to the listener's context, inefficient in use and lacking the naturalness that comes from rapid appreciation of the speaker-listener environment.  A key scientific challenge is to better understand how "talker-listeners" respond to context and to apply these findings to the modification of natural (live/recorded) and generated (synthetic) speech. The ISCA-supported Listening Talker (LISTA) workshop brings together linguists, psychologists, neuroscientists, engineers and others working on human and machine speech perception and production, to explore new approaches to context-sensitive speech generation.

The workshop will be single-track, with invited talks and contributed oral and poster presentations. An open call for a special issue of Computer Speech and Language on the theme of the listening talker will follow the workshop.

Contributions are invited on any aspect of the listening talker, including but not limited to:

- theories and models of human communication involving the listening talker
- human speech production modifications induced by noise
- speech production changes with manipulated feedback
- algorithms/vocoders for speech modification
- transformations from casual to clear speech
- characterisation of the listening context
- intelligibility and quality metrics for modified speech
- application to natural dialogues, PA, teleconferencing

Invited speakers

Torsten Dau  (Danish Technical University)
Valerie Hazan  (University College, London)
Richard Heusdens  (Technical University Delft)
Hideki Kawahara  (Wakayama University)
Roger Moore  (University of Sheffield)
Martin Pickering  (University of Edinburgh)
Peter Vary  (Aachen University)
Junichi Yamagishi  (University of Edinburgh)
	
Important dates

30th January 2012: Submission of 4-page papers 
27th February 2012: Notification of acceptance/rejection

Co-chairs

Martin Cooke  (University of the Basque Country)
Simon King  (University of Edinburgh)
Bastiaan Kleijn  (Victoria University of Wellington)
Yannis Stylianou  (University of Crete)


-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.