Neat alternative it seems. I do wish there was something like this that was on-device only. My guess is maybe there is, with some local LLM, but I’m just unaware of it.
I wrote a very basic version of this for automatically removing ads from podcasts. It works exactly as I hoped, but needs a little work yet before sharing. Very basic combination of Whisper -> Gemini -> FFMPEG. Gemini is so good and so cheap (pennies) that I haven't prioritized using local models, but that should work too.
Neat alternative it seems. I do wish there was something like this that was on-device only. My guess is maybe there is, with some local LLM, but I’m just unaware of it.
I wrote a very basic version of this for automatically removing ads from podcasts. It works exactly as I hoped, but needs a little work yet before sharing. Very basic combination of Whisper -> Gemini -> FFMPEG. Gemini is so good and so cheap (pennies) that I haven't prioritized using local models, but that should work too.
Thanks! I can see how on-device can be appealing, especially for the editor with a local model transcribing the media files to text.