feature request: auto-transcribe audio files

[andrewheusser]

I could see this working in two ways:

1) a function that takes an audio file and returns a transcript. this could be based on google speech, or our own speech model.

2) a higher level wrapper function that takes a group of encoding lists and audio files, and creates a pyro object for further analysis

[jeremymanning]

this would be awesome. if we go the google cloud speech route, we'll need to add setup instructions too (e.g. to show users how to set up google cloud)

[andrewheusser]

i think we should offer google cloud for sure, because it will be pretty easy for me to add at this point, and then attempt to build our own 'open-source' model, which comes with limited guarantees :)

[andrewheusser]

does anyone that you know in the dept, or cs dept work on speech recognition? @jeremymanning

[jeremymanning]

possibly X-D Yang? http://web.cs.dartmouth.edu/people/xing-dong-yang