Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

DiSCo - a German evaluation corpus for challenging problems in the broadcast domain

: Baum, D.; Schneider, D.; Bardeli, R.; Schwenninger, J.; Samlowski, B.; Winkler, T.; Köhler, J.

Postprint urn:nbn:de:0011-n-1322830 (321 KByte PDF)
MD5 Fingerprint: 706ceee4aad193255d3e304206986125
Erstellt am: 1.6.2010

European Language Resources Association -ELRA-, Paris:
LREC 2010, 7th International Conference on Language Resources and Evaluation. Proceedings : Workshops & Tutorials May 17-18, May 22-23, Main Conference May 19-21, Valletta
Paris: ELRA, 2010
ISBN: 2-9517408-6-7
International Conference on Language Resources and Evaluation (LREC) <7, 2010, Valletta>
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IAIS ()

Typical broadcast material contains not only studio-recorded texts read by trained speakers, but also spontaneous and dialect speech, debates with cross-talk, voice-overs, and on-site reports with difficult acoustic environments. Standard approaches to speech and speaker recognition usually deteriorate under such conditions. This paper reports on the design, construction, and experimental analysis of DiSCo, a German corpus for the evaluation of speech and speaker recognition on challenging material from the broadcast domain. One of the key requirements for the design of this corpus was a good coverage of different types of serious programmes beyond clean speech and planned speech broadcast news. Corpus annotation encompasses manual segmentation, an orthographic transcription, and labelling with speech mode, dialect, and noise type. We indicate typical use cases for the corpus by reporting results from ASR, speech search, and speaker recognition on the new corpus, thereby obtaining insights into the difficulty of audio recognition on the various classes.