ixd: Getting speech recognition from p5.speech into Max

Posted on
itpnyu ixd interaction_design_studio websockets node

Got a basic functional prototype working for the microphone team.

We were looking for a method to use the p5 speech library for continuous speech recognition and getting the results to Max.

I found a solution after Luke duBois suggested using websockets.

  1. Node server hosts p5 sketch
  2. Browser sends word back via websocket
  3. Node server writes word to text file
  4. Max gets text file via GET request very often

Next Steps

The worst part is the continuous GET requests from Max. I’ll see if I can replace this with sending messages via OSC.