Speech To Text Transcription via Google's S2T Engine. Uses LIAM for diarization and segmentation metadata. Uses ffmpeg for audio tasks.
Original idea from here: http://www.raspberrypi.org/phpBB3/viewtopic.php?uid=19416&f=38&t=27290&start=0
API v2 script from here: https://github.com/gillesdemey/google-speech-v2
LIUM Diarization based on this script: http://www-lium.univ-lemans.fr/diarization/doku.php/single-show_diarization
##Requires:##
OSX 10.6.8+
Java CLI Runtime (Available here)
It may run on other *nix systems, but I can't promise anything.