On Mon, 22 Oct 2012 12:47:51 -0700 Carlos Alvarez <car...@televolve.com> wrote: > > In-house transcriptions are definitely out of the question, but any > experience with outsourced solutions would be useful. As far as I can tell > the current service is automated, and as awful as Google Voice, yet they > find it useful. Their existing carrier uses Broadsoft and I'm not sure if > they have that built in. >
Voice recognition for asterisk based on Google speech API is already available[1], the problem with this service is that it's limited to 20-30 seconds of speech data, which isn't suitable for transcripting voicemails. If you are able to find a reliable way of chopping speech samples in segments no bigger than 20 seconds based on silence detection, so words wont be cut in half, you might come up with something very similar to Google Voice transcription service. But I would recommend against using this into production since google haven't yet defined the terms of service for speech recognition, and its more or less a hack for now. [1] http://zaf.github.com/asterisk-speech-recog/ ---------------- Lefteris Zafiris -- _____________________________________________________________________ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users
