<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
<h2><span class="titolo1"><a href="http://www.tc-star.org"><img
src="cid:part1.03070305.09010308@elda.org" border="0" height="64"
width="70"></a></span></h2>
<h2><span class="titolo1">TC-STAR Third Evaluation Campaign 2007 </span></h2>
<h3>Call for Participation </h3>
<table height="828" width="781">
<tbody>
<tr>
<td>
<p class="body" align="justify"> <br>
TC-STAR is an European integrated project focusing on Speech-to-Speech
Translation (SST). To encourage significant advances in all SST
technologies, annual competitive evaluations are organized. Automatic
Speech Recognition (ASR), Spoken Language Translation (SLT) and
Text-To-Speech (TTS) are evaluated independently and within an
end-to-end system. The project targets a selection of unconstrained
conversational speech domains—speeches and broadcast news—and three
languages: European English, European Spanish, and Mandarin Chinese. </p>
<p class="bodygrande"><strong>The second evaluation campaign took
place in March 2006 and was concluded by an <a
href="http://www.elda.org/tcstar-workshop_2006">international workshop</a>.
TC-STAR welcomes outside participants in its 3rd
evaluation of January-February 2007. This participation is free of
charge. </strong></p>
<p class="body" align="left">The TC-STAR 2007 evaluation campaign
will consider:</p>
<div align="justify"><span class="body"><strong>SLT in the
following directions<br>
</strong>a) Chinese-to-English (Broadcast News)<br>
b) Spanish-to-English (European Parliament plenary speeches; Cortes
Spanish Parliament speeches)<br>
c) English-to-Spanish (European Parliament plenary speeches)<br>
<br>
The SLT evaluation of European Parliament plenary speeches will cover
three types of input:<br>
1. ASR output (single best sentence, and more: N-best, word lattices)<br>
2. verbatim transcription,<br>
3. final text edition.<br>
</span>
<div align="justify">
<p> <span class="body"><strong>ASR in the following languages<br>
</strong>a) English (European Parliament plenary speeches)<br>
b) Spanish (European Parliament plenary speeches; Cortes Spanish
Parliament speeches)<br>
c) Mandarin Chinese (Broadcast News)</span></p>
<span class="body"><strong>TTS in Chinese, English, and Spanish
under the following conditions:<br>
</strong>a) Complete system: participants use their own training
data<br>
b) Voice conversion intralingual and crosslingual, expressive speech:
data provided by TC-STAR<br>
c) Component evaluation</span></div>
</div>
<p class="body" align="justify">For
ASR and SLT, training data will be made available by the TC-STAR
project for English and Spanish and can be purchased at LDC for
Chinese. Development data will be provided by the TC-STAR project.
Legal issues regarding the data will be detailed in the 2nd Call For
Participation.</p>
<p class="body" align="justify">All participants will be given
the opportunity to present and discuss their results in the 3rd TC-STAR
evaluation<strong> workshop in Aachen in March 2007</strong>. </p>
<p class="body"><br>
<span class="bodygrande">Tentative schedule:<br>
</span>Registration: October 2006 (early expression of interest
is welcome)<br>
ASR evaluation: from 21st of January to 28th of January 2007<br>
SLT evaluation: from 31st of January to 7th February 2007<br>
TTS evaluation: from 9th of February to15th of February 2007<br>
<br>
Workshop: March 28-30, 2007 in Aachen, Germany </p>
<p class="body"><br>
Contact: Djamel Mostefa (ELDA) <br>
e-mail: mostefa at elda.org <br>
tel. +33 1 43 13 33 33</p>
</td>
</tr>
</tbody>
</table>
</body>
</html>