<%@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
  • 2002 Speaker Recognition Evaluation
    Extended Data

    Speaker Conversation Table

    The format of the speaker conversation table is defined in section 8.3.1 of the evaluation plan.

    Evaluation Control File [ NEW March 21st, 2002 ]

    The official evaluation control file (version 2) is a subset of the originally released control file. This control file has 4513 models, 59,965 trials, and 652 speakers.

    Sites may still process version 1 if they wish, and NIST will subset the submision file for comparitive results. This control file has 10,933 models, 156,184 trials, and 1065 speakers.

    The format of the evaluation control file is defined in section 8.3.2 of the evaluation plan.

    ASR Transcripts available for the extended data evaluation

    These ASR transcripts were generated at NIST, using a version of BBN's Byblos system. This was a near real time system which was NOT optimized for this task, and therefore the word error rate is expected to be much higher than it was for the transcripts provided for last year's dry run evaluation.

    Development ASR Transcripts

    There was a request for sample transcripts produced by the Byblos recognizer NIST is using for the evaluation. NIST has processed the Original Switchboard corpus with this recognizer.



     



    [ SRE2002 ]

     

     

    Page Created: Month Day, Year
    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