Speaker Diarization / Recognition

Description
<p><a href="https://en.wikipedia.org/wiki/Speaker_diarisation">Speaker diarization</a> is the process of splitting 1 input audio stream into separate output audio streams according to speaker identity. This technology has uses in broadcasting and speech recording.<br/></p><p><b>Potential Implementations:</b><br/><a href="http://www-lium.univ-lemans.fr/diarization/doku.php/welcome">http://www-lium.univ-lemans.fr/diarization/doku.php/welcome<br/></a><a href="http://mistral.univ-avignon.fr/index_en.html">http://mistral.univ-avignon.fr/index_en.html</a> or <a href="https://github.com/rampa069/LIA_RAL/tree/master/LIA_SpkSeg">https://github.com/rampa069/LIA_RAL/tree/master/LIA_SpkSeg </a>(has not been updated since ~2012)<br/><br/><b>Desired Parameters:</b> Input URL, Output URL<br/><b>Expected input:</b> URL to audio file<br/><b>Expected output:</b> Collection of 1 or more audio files written to a data collection URL</p><p><b>Note:</b> The ideal implementation of this algorithm will allow for seamless integration with the <a href="https://algorithmia.com/algorithms/sphinx/SpeechRecognition">Speech Recognition algorithm by Sphinx</a><br/></p>
Discussion
  • {{comment.username}}
Status
Active
submission(s) pending review
Bounty expires in
Bounty expired
Bounty
75
Tags
(no tags)