IARPA Babel Mongolian Language Pack IARPA-babel401b-v2.0b

Introduction
<br>
<br>IARPA Babel Mongolian Language Pack IARPA-babel401b-v2.0b was developed by Appen for the IARPA (Intelligence Advanced Research Projects Activity) Babel program. It contains approximately 204 hours of Halh Mongolian conversational and scripted telephone speech collected in 2014 along with corresponding transcripts.
<br>
<br>The Babel program focused on underserved languages and sought to develop speech recognition technology that could be rapidly applied to any human language to support keyword search performance over large amounts of recorded speech.
<br>Data
<br>
<br>The Halh Mongolian speech in this release represents that spoken by native speakers in Mongolia. The gender distribution among speakers is approximately equal; speakers' ages range from 16 years to 61 years. Calls were made using different telephones (e.g., mobile, landline) from a variety of environments including the street, a home or office, a public place, and inside a vehicle.
<br>
<br>Audio data is presented as 8kHz 8-bit a-law encoded audio in sphere format and 48kHz 24-bit PCM encoded audio in wav format.
<br>
<br>Transcripts are encoded in UTF-8 in both Mongolian Cyrillic and a romanization scheme developed by Appen; they cover approximately 77% of the speech data. Further information about transcription methodology is contained in the documentation accompanying this release.
<br>
<br>Evaluation data is available from NIST in support of OpenKWS.
<br>Samples
<br>
<br>Please view the following samples:
<br>
<br> Audio Sample (SPH)
<br> Transcript Sample (TXT)
<br> Romanized Transcript Samples (TXT)
<br>
<br>Updates
<br>
<br>None at this time.
<br>Copyright
<br>Portions © 2015 U.S. Government
<br>
<br>The U.S. Government acquired this data from Appen Pty Ltd, which assigned the copyright to the data to the U.S. Government.