Oooh, I see! But it could be done through an AI that uses the stream's sound as its data input for the transcription, right? A voice-to-text model.
You are viewing a single comment's thread from:
Oooh, I see! But it could be done through an AI that uses the stream's sound as its data input for the transcription, right? A voice-to-text model.
It could. But it also costs significantly more to process audio than text. It's simply not economically viable, at least not for me.
Understood!