• If you never picked a speech engine, Hedy now offers
    Improved transcription
    and downloads the recommended model for your language in the background, with a progress bar. It switches over once the download is complete.
  • Starting a session pauses the download, and it picks up where it left off afterwards. If a download doesn't finish, Hedy keeps working as before.
  • Speech model downloads from
    Settings → Speech & AI
    also wait while a session runs, pick up afterwards, and can be cancelled.
  • A speech engine you picked yourself is never changed.