PocketSphinx+PG+JavaScript Voice/Text Experiment

#1

Post by vmorgulis » Sat, Feb 25, 2017, 3:52 PM UTC

PocketSphinx+PG+JavaScript Voice/Text Experiment
vmorgulys.github.io

Re: PocketSphinx+PG+JavaScript Voice/Text Experiment

#2

Post by vmorgulis » Sat, Feb 25, 2017, 3:53 PM UTC

Original video: https://www.youtube.com/watch?v=0KR2MSFROLI

CMUSphinx: http://cmusphinx.sourceforge.net/

Re: PocketSphinx+PG+JavaScript Voice/Text Experiment

#3

Post by detaro » Sat, Feb 25, 2017, 3:59 PM UTC

vmorgulis wrote: post #2

Original video: https://www.youtube.com/watch?v=0KR2MSFROLI CMUSphinx: http://cmusphinx.sourceforge.net/

So, what am I looking at? It seems like you fed the audio in PocketSphinx to get time-tagged text and the site basically shows said text as subtitles to what was said, is that the gist of it?

Re: PocketSphinx+PG+JavaScript Voice/Text Experiment

#4

Post by vmorgulis » Sat, Feb 25, 2017, 4:53 PM UTC

detaro wrote: post #3

vmorgulis wrote: post #2

Original video: https://www.youtube.com/watch?v=0KR2MSFROLI CMUSphinx: http://cmusphinx.sourceforge.net/

So, what am I looking at? It seems like you fed the audio in PocketSphinx to get time-tagged text and the site basically shows said text as subtitles to what was said, is that the gist of it?

> ... is that the gist of it?

Yes, it is.

I'd like to improve the speech recognition and expected some advice about that.

Another possibility is to add a semantic level with NLP or use another library like Kaldi (http://kaldi-asr.org/).

Another particularity: the WAV file is serialized in JSON (as an array).