Iraqi Arabic Conversational Telephone Speech, Transcripts

<h3>Introduction</h3><br>
<p>Iraqi Arabic Conversational Telephone Speech, Transcripts was developed by Appen Pty Ltd, Sydney, Australia and contains transcripts for roughly 3000 mins of speech from Iraqi Arabic speakers taking part in spontaneous telephone conversations in Colloquial Iraqi Arabic.</p><br>
<p>This corpus was collected and transcribed in 2003 and 2004 by Appen Pty Ltd, Sydney, Australia. The corresponding speech files for these transcripts are available in <a href="../../../LDC2006S45">Iraqi Arabic Conversational Telephone Speech (LDC2006S45)</a>.</p><br>
<h3>Data</h3><br>
<p>A total of 478 conversation sides from 474 unique speakers are provided, and most of these transcripts contain both sides of a conversation (202 transcripts with both sides and 74 with just one side). The average duration per call is about six minutes, so each call side contains about three minutes of speech, on average.</p><br>
<p>The transcripts directory contains three subdirectories:</p><br>
<p style="padding-left: 30px;">devtest -- 12 two-sided transcripts<br />train1c -- 74 single-sided transcripts<br />train2c -- 190 two-sided transcripts</p><br>
<p>The single-sided transcripts represent just one side of a normal conversation with the speaker info (A or B) included in the filename, formulated like such: AR-[DATE]-[TIME]_[LINE]-[A or B].txt</p><br>
<p>The "devtest" set represents a relatively balanced (representative) sample drawn from the total pool of collected calls, based on a test-set selection process applied by the National Institute of Standards and Technology (NIST) and based on demographic, phone and audit information as provided by Appen.</p><br>
<h3>Samples</h3><br>
<p>For an example of the transcripts in this release, please review this <a href="desc/addenda/LDC2006T16.txt" rel="nofollow">sample</a>.</p><br>
<h3>Updates</h3><br>
<p>None at this time.</p></br>
Portions © 2006 Trustees of the University of Pennsylvania