dejavu
dejavu copied to clipboard
worldveil
AAC Direct input
Is there a way that I can feed a live AAC audio stream into dejavu for recognition? I thought about saving part of the stream to file and then running file recognize but I don't really want to add the time needed to save them then recognize. I need this to be almost as realtime as possible. I also thought about looping the audio back into the mic input somehow so I wouldn't have to change anything, but I thought it would be better to do that.
I've never played with pyaudio before, I'm trying to figure out where the audio goes into dejavu, but I'm not getting very far with that.
Would be a lovely feature. But don't know how to configure this either.
it depends on the ffpmeg extracting the AAC files to wav format. so you have to check that first
you also have to match the sample rates for the recorded files refer to issue #62 to know more about