QuietSteno

Home / Tools / Translate audio to English

Translate audio or video into English text

Have a recording in Spanish, German, Japanese or another language? Get it as English text or English subtitles in one step, processed on your own device.

Drop an audio or video file here

MP3, WAV, M4A, MP4, MOV, WebM, OGG or FLAC. The file stays on this device.

Runs on your device with the open-source Whisper model. Nothing is uploaded. How privacy works

To translate audio to English, add the recording to QuietSteno, select the language spoken in it and set Output to “Translate to English”. The Whisper speech model listens to the original language and writes English text directly, on your device. Download the result as TXT, or as English SRT or VTT subtitles.

How to translate a recording into English

  1. Add the audio or videoAny supported file works: MP3, M4A, WAV, MP4 and others.
  2. Select the spoken languageChoose the language that is actually spoken, for example Spanish or Japanese.
  3. Set Output to “Translate to English”This tells the model to write English instead of the original language.
  4. ExportDownload English text, or SRT or VTT to add English subtitles to the video.

How speech translation works here

The Whisper model was trained to do two jobs: write down speech in its own language, and translate speech from other languages into English. Translation happens in one pass from the audio, without producing a transcript in the original language first.

The direction is fixed: any supported language into English. To get text in a different target language, transcribe the recording in its original language and paste that transcript into a text translator.

What to expect from the translation

The result reads as natural English and conveys the meaning well for clear, everyday speech. It is a looser rendering than a professional translation: idioms may be simplified, and names or specialist terms can be wrong.

  • Use it to understand a recording, to draft English subtitles, or to decide whether something is worth a human translation.
  • Use Most accurate. Translation benefits more from the larger model than plain transcription does.
  • Do not rely on it alone for legal, medical or published material. Have a fluent speaker review it.

Questions and answers

Which languages can be translated into English?

The 26 non-English languages in the language menu, including Spanish, French, German, Italian, Portuguese, Dutch, Polish, Russian, Ukrainian, Turkish, Arabic, Hindi, Japanese, Korean and Chinese.

Can it translate English audio into another language?

No. The model translates into English only. Transcribe the English audio first, then use a text translator for the language you need.

Is my recording uploaded to a server?

No. The speech model is downloaded to your browser and runs on your own processor or graphics chip. The recording is read from your disk by the page and never sent anywhere.

Can I get both the original transcript and the English translation?

Yes, by running the file twice: once with Output set to “Same language” and once with “Translate to English”. The “Run again with these settings” button reuses the same file.

Are the English subtitles timed to the original speech?

Yes. Each translated line keeps the time at which the original words were spoken, so English SRT or VTT files line up with the video.

Related

Last updated 13 November 2026.