Comment by icelancer

it does this at a variable rate of frames which you can set - not sure if it is transcribing or natively understanding audio, but I think it's the latter since it is much faster than most transcription models I am aware of

Regardless you are right - I can roll my own.... but why