Speech to Text

[ritwik12]

Virtual assistant is only text to speech at this moment as it is hard to implement Speech to text in C language.
But if we can have Speech to text too, then we'll be able to use it with voice commands also and that'll be an huge improvement.

[speedy-software-ch]

isn't there a library to do that?

[ritwik12]

@speedy-software-ch Non of the APi's are available for C.
or maybe I couldn't find any.

[speedy-software-ch]

ok. i'll try to research it too.

[ritwik12]

@speedy-software-ch Thanks

[KeenanTaggart]

I found a couple of good options for this although I haven't tested them myself.

Carnegie Mellon maintains speech-recognition software in a suite called CMU Sphinx. PocketSphinx is a version written in C and is meant to have low overhead for embedded systems (I think it would be cool to be able to run Virtual Assistant on a headless computer like a Raspberry Pi with a speaker/mic attached). The homepage for CMUSphinx is here and the downloads page links to PocketSphinx specifically.

There's also this option here, which seems to be a C wrapper for the Google Speech API. It might be simpler to set up/use but would require a constant web connection.

[ritwik12]

@KeenanTaggart I got your idea of using headless computer, but virtual assistant uses a lot of Linux functionalities. So it will be a problem and we need to write lots of functionalities again.

Also, there is no problem with constant web connection as many of the features uses internet and to use virtual assistant a web connection is always needed.