Elektrine lite

← Feed

@jaybrunet@podcastindex.social

Post #2382479

2026-04-05 05:32 UTC

And I'm realizing my agent can help me streamline a workflow like this... Random noisy voice notes -> deep-filter -> transcription -> lots of text -> AI agent processing (whatever Openrouter model) -> grammatically perfect, coherent text -> TTS output.

Replies (1)

  • Almost seems like an elaborate form of procrastination, but in the long run I think it will work, because I'm not as critical about the input recordings, and I'm confident the output will make more sense (in less time) than if I recorded everything perfectly using my own voice. I'm usually recording while hiking and that's just a pain to edit afterwards -- I could use Descript or https://github.com/bugbakery/audapolis -- but I believe it'll be easier to use TTS, even if it doesn't sound as err natural ;-)

    Open ##2382480