Special Session

Call for Participation: The NATO non-native Military Air Traffic Communications database (nnMATC)

Novel techniques for the NATO non-native Air Traffic Control and HIWIRE cockpit databases

Following a series of special interest sessions and (satellite) workshops, at Lisbon (1995), Leusden (NL, 1999) and Aalborg (2001), the NATO research task group on speech and language technology, RTO IST031-RTG013, organizes a special session at Interspeech 2007. After having studied various aspects of speech in noise, speech under stress, and non-native speech, the research task group has been studying the effects of all of these factors on various speech technologies.

To this end, the task group has collected a corpus of military Air Traffic Control communication in Belgian air space. This speech material consists predominantly of non-native English speech, under varying noise and channel conditions. The database has been annotated at several levels:

• word transcriptions, which allow research to be conducted on automatic speech recognition and named entity extraction,
• Speaker turns, identified by call signs, allowing for research in speaker recognition and clustering and tracking of conversations.

The NATO research task group is making this annotated speech database available for speech researchers, who want to develop novel algorithms for this challenging material. These new algorithms could include noise-robust speaker recognition, robust speaker and accent adaptation for ASR, and context driven named entity detection. In order to facilitate a common task, we have written a suggested test and evaluation plan to guide researchers. At the special session we will discuss research results on this common data set.

Updated Transcriptions, Readme and database description

• [15 Mar 2007] (new) A pre-print of the Interspeech article we intend to submit, describing the rationale for the database and some statistics thereof.
• [12 Mar 2007] Updated version: new sei-lei transcription files) Raw transcription material, in various formats for the various experiments proposed, can be found here (1.0 MB file).
• [5 Mar 2007] A README file is available now.

Data order form

The data ordering period is over; please contact us at Interspeech 2007. Really sick people used the form to send us spam.

Submission

Researchers who are interested in contributing to this special session are invited to submit a paper according to the regular submission procedure of INTERSPEECH 2007, and to select `Novel techniques for the NATO non-native Air Traffic Control database' in the special session field of the paper submission form. The paper submission deadline is 23 March 2007.

Contact

Co-organizer: Alex Potamianos (Technical University of Crete)