1 comments

  • glashatay 5 hours ago

    the retry with saved audio is a solid UX choice. most STT tools just discard the failed attempt. curious what backend you use for the transcription - local model or API? the private aspect suggests local but wanting to confirm