<%@LANGUAGE="JAVASCRIPT" CODEPAGE="65001"%> NIST Speech Group Website
Information Technology Lab, Information Access Division NIST: National Institute of Standards and Technology


  • Multimodal Information Group Home
  • Benchmark Tests
  • Tools
  • Test Beds
  • Publications
  • Links
  • Contacts
  • Story Segmentation Task

    Story segmentation is the task of segmenting the stream of data from a source into topically cohesive stories. Since text (newswire) sources are supplied in segmented form, this task applies only to the audio subset of the TDT corpus (radio and TV). Segmentation of audio signals may be performed using the audio signal itself or the provided manual/automatic textual transcriptions of the audio signal.

    The graphic on the bottom depicts the stream of data as a sequence of stories and non-stories. Segmentation systems place boundaries between these units and the evaluation software measures the system's ability to do so.

    Story segmentation performance depends on the form of the source and on the maximum time allowed before segmentation decisions must be output. These factors are taken into account for the evaluation by defining appropriate experimental control conditions.

    For more information concerning the current research in TDT segmentation, there were two system description papers submitted for the 1999 TDT workshop. Also, the TDT Research Links contains pointers to previous year's TDT segmentation research.

     

     

    Page Created: August 21, 2007
    Last Updated: November 4, 2008

    Multimodal Information Group is part of IAD and ITL
    NIST is an agency of the U.S. Department of Commerce
    Privacy Policy | Security Notices|
    Accessibility Statement | Disclaimer | FOIA