(059) previous ~ index ~ next

To: tdt-distrib@unagi.cis.upenn.edu
From: George Doddington <doddington@nist.gov>
Subject: TDT3 source conditions
Date: Mon, 29 Mar 1999 19:26:43 -0500

There will be three official conditions for the TDT3 source data:

* newswire text and a TDT3-supplied manual transcription
for the audio sources

* newswire text and a TDT3-supplied ASR transcription
for the audio sources

* newswire text and the sampled data signal
for the audio sources

Each research site is encouraged to study as many of these conditions
as may be productively done. However, all sites that perform the task
are required to perform at least one evaluation under a REQUIRED source
condition. This is the second condition, namely the newswire text and
a TDT3-supplied ASR transcription for the audio sources.

An automatic MT translation of the Mandarin language data into English
will also be supplied, as an OPTIONAL form of the Mandarin source data.
This will be derived from the source data for the required condition
and therefore satisfies the required source data condition. Story
boundary information for the MT translation will also be supplied.
--
George Doddington at NIST: doddington@nist.gov or 301/975-3261
(059) previous ~ index ~ next

Last updated Thu May 13 09:28:21 1999