• DocumentCode
    2181378
  • Title

    Robust speaker turn role labeling of TV Broadcast News shows

  • Author

    Damnati, Géraldine ; Charlet, Delphine

  • Author_Institution
    Orange Labs., France Telecom, Lannion, France
  • fYear
    2011
  • fDate
    22-27 May 2011
  • Firstpage
    5684
  • Lastpage
    5687
  • Abstract
    Speaker role recognition in TV Broadcast News shows is addressed in this paper with a particular focus on speaker turn role labeling. A mixed approach combining speaker clustering and analysis of Automatic Speech Recognition output is proposed for assigning speaker turns a role among: anchor, reporter and other. 86% classification accuracy is obtained for automatically segmented speaker turns on a 6.5 hours test corpus of 14 TVBN shows mixing news and conversational speech.
  • Keywords
    speaker recognition; speech recognition; television broadcasting; TV broadcast news shows; automatic speech recognition; robust speaker turn role labeling; speaker clustering; time 6.5 hr; Accuracy; Feature extraction; Labeling; Speech; Speech processing; Speech recognition; Training; Broadcast News and Conversation speaker distillation; Speaker role; spoken language understanding;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
  • Conference_Location
    Prague
  • ISSN
    1520-6149
  • Print_ISBN
    978-1-4577-0538-0
  • Electronic_ISBN
    1520-6149
  • Type

    conf

  • DOI
    10.1109/ICASSP.2011.5947650
  • Filename
    5947650