Omega2 voice recognition



  • Hello everybody, I am about to build a voice recognition solution and I want to try omega2.
    I installed alsa and I am thinking about using sphinx (http://cmusphinx.sourceforge.net) or Julios for recognition and festival for text to speech.
    Do you have any suggestions how to make this happen? Probably I need to build them from source. I dont know if it is ok with omega. I dont know if omega2 is suitable for this.
    Thanks for your advices, comments



  • Have a look at the jasper project

    http://jasperproject.github.io/documentation/



  • Yeah, I know about jasper, but I think this would not be possible, because packages needed for this are not available for omega. And you can't just make "make".



  • Cross compiling is indeed a bit of an involved project, but generally is for any embedded system until you get the hang of it.

    You might also look at the similarly MT7688-based Seeed Respeaker thing - although they are using a different starting point and a different C library, it's likely a full-scale replacement of the Onion system with theirs could be made to work on the Omega2. Unclear if their hardware will ever ship, but starting to look like they have some repositories up. You'd need a version for the base board, rather than that using their add-on directional microphone array.


Log in to reply
 

Looks like your connection to Onion Community was lost, please wait while we try to reconnect.