Combine betty with speech to text? about betty

[pickhardt]

Not yet, as far as I know.

[deltaidea]

There's this related project: https://github.com/debugger22/Jarvis

Anyway, I don't think it's a good idea since it's quite difficult to dictate paths, usernames, etc. Of course, you can combine text and speech, but:

Betty is a friendly English-like interface for your command line.

It solves the problem quite well. Why expanding and generalizing?
Or maybe I just want to argue pointlessly...

[lpanebr]

I agree that this fall out of the scope of Betty. There are other projects for that. Also I believe that OSs will eventually support that natively.

Betty should should stay on track and focus in understanding natural text input and solving user requests while teaching them the commands behind her brain.

[pannous]

It would be only 10 lines of code though:
https://github.com/gillesdemey/google-speech-v2
maybe in separate listen.sh

[pannous]

actually three lines:

`rec -rate 44100 out.flac silence 1 0.50 0.1% 1 2.0 0.1%`
result=`curl -X POST --data-binary @out.flac --header 'Content-Type: audio/x-flac; rate=44100;' 'https://www.google.com/speech-api/v2/recognize?&lang=en-us&key=AIzaSyCnl6MRydhw_5fLXIdASxkLJzcJh5iX0M4'`
parse result["result"][0]["alternative"][0]["transcript"]

@gillesdemey : uhm, did you post your google api key on git??

[gillesdemey]

Nope, the key was extracted from the Google Hotwords Chrome extension and has no courtesy limit :)

Jarvis also uses the same key, and a bunch of other STT projects so speak away!

[pannous]

phew ;}
thanks!