Hi
I think Google's speech to text facilities (Google Voice automatic transcription of voicemail, automatic captioning of videos on YouTube etc) are quite impressive.
I did look to see whether Google has made it available through an API and it seems they haven't (not that I blame them!). A cloud computing service providing speech to text functionality would be pretty cool though.
Is there some sort of "hack" that I can use to access the speech to text. My architecture basically comes down to this - a short 15-20 second wav/mp3/other clip as the input, output is plaintext.
Any ideas people?