Hi, I'm trying to create a stream of PyAudio to make a speech-to-text on a website (with a microphone).
I'm not really sure what the best way of achieving this so feel free to suggest new ways.
The current Structure is: WEB -> Api gateway (web socket) -> EC2 instance that have model loaded for the STT job.
I want to get back to the WEB the results of the streaming but I keep getting that error: "OSError: [Errno -9996] Invalid input device (no default output device)"
I think the problem is that the PyAudio code is on the remote server and it doesn't have a microphone. is it possible to send chunks from the web (where we do have a microphone) to the PyAudio stream?
Any suggestions?
