Live Audio recognition

I was working on Reconocimiento de audio simple: reconocimiento de palabras clave  |  TensorFlow Core and I completed it, but I was wondering how I could do this recognition live instead of on an audio file. I’m not very experienced, so help would be appreciated. Also, are there any good Tensorflow tutorials on live speech recognition? Thanks.

Hi Henry,

In which environment you want to capture the audio? on a phone or browser?

The idea is, that model could be ported to the place where the audio is being captured and then you’d need to keep feeding the audio to the model and getting the results
So, in summary the capture and feeding the data to the model is just regular coding

That makes sense. How would I go about doing that? I am working on Ubuntu 20.04, and my language is Python, so I do not know if a browser can send data to python, so should I use a certain library?
I apologize for the question. I am not very experienced. Also, you guys are the only community that responded to me. :smiley:

Hey Henry, try this or google similar terms to those, you’ll find a lot of ways to get audio from your mic in real time. Hope it works!

1 Like