Closed michaelwarren1106 closed 9 years ago
Hi.
I'm assuming you are talking about dictation vs voice commands right? The speech-to-text always gives you an hypothesis of an audio speech represented as text, so there is always a dictionary involved.
CMU Sphinx can handle both cases, dictation and voice commands (as dictation with a small dictionary), so TLSphinx also has this capabilities. You can check the basic usage tutorial on how to use Pocketsphinx for a simple explanation. More documentation can be found in the full Pocketsphinx API.
Also check the TLSphinx tests cases for an example of how to use this APIs in Swift.
Hope this is what you ask for :sweat_smile:
Yes, i'm talking about dictation, not voice commands :) Thanks for the help!
I'm trying to get a speech-to-text feature up and running in a sample app. What i'm looking for is full hypothesis speech-to-text capability, instead of just keyword dictionary matching. Does this library accomplish that? If so, is there any tutorial on how?