Transcribe files

Turn recordings and videos into text and subtitles on your own computer. Open Transcribe a file in the sidebar.

Add files

Drop audio or video files anywhere on the HushWise window, or press Choose files. MP3, M4A, WAV, MP4, MOV and more work. The audio is decoded in memory and never copied.

Choose a format

Under Save as, choose at least one format:

Format Use
.txt Plain text
.md Markdown
.srt Subtitles
.vtt Subtitles

You can also pick the language, or turn on Translate to English where your models support it. See Languages and translation.

The queue

Files wait in a queue and are transcribed one by one in the background. Dictation always comes first: files pause while you are dictating and carry on afterwards. You can close the window. Files keep transcribing while HushWise runs. Clear finished removes completed items from the list.

Where transcripts are saved

By default, each transcript is saved next to its file. To use one folder instead, choose it under Save in. Use each file’s folder goes back to the default. Show in folder opens the result.

If a file fails

  • “Not an audio or video file”: choose a different file.
  • “This file couldn’t be read”: it may be damaged, or have no sound.
  • “No speech model”: download one on the Models page.
  • “Couldn’t save the transcript”: choose another folder.

File transcription pauses if your trial has ended. See Licence and trial.