NewSee any caption style on your own video before it's captioned
CAPWAV
Video to text

Transcribe videoto text.

Upload a video and get its speech as text, timed to every word. Correct anything in the editor, then download a plain-text transcript or a subtitle file.

Caption style
All 358 →

Clean · you'll see it on your own video before captions are made. Only the audio is sent to be transcribed.

workout.mp4Subtitles readySpoken language: English
  1. 0:00.2Okay, so today we're doing
  2. 0:02.2a full-body warm-up.
  3. 0:03.5Keep your core tight
  4. 0:05.1and breathe out slowly.

A transcript in minutes

CAPWAV transcribes speech in 90+ languages, word by word, with each word's timing. Choose the spoken language for the best results, or let it detect the language.

Use the text for show notes, blog posts, quotes or translation.

Edit before you download

Fix names and terms by clicking a word. Your corrections carry into every download and into the captions.

Your video stays on your device: only the audio is sent to be transcribed.

How it works

How to transcribe video to textin three steps.

  1. Step 01

    Upload your video

    Add a clip from your phone or computer and choose its spoken language.

  2. Step 02

    Transcribe and check

    CAPWAV writes the transcript. Click any word to correct it.

  3. Step 03

    Download

    Download TXT, SRT or VTT — free on every plan.

FAQ

Questions,answered.

Something else? Contact us

How long can the video be?

Videos with up to about 15 minutes of speech can be transcribed today.

Is it free?

The Free plan includes 2 minutes of transcription a month, and TXT, SRT and VTT downloads are free on every plan.

Which languages can it transcribe?

90+ languages, including English, Spanish, Portuguese, French, German, Hindi, Arabic and Japanese.

Does my video get uploaded?

No. Only the audio is sent to be transcribed; the video stays on your device.

Your next reel is
one upload away.

Start free with 2 minutes of captions a month. Upgrade when you're posting more.