Turn a video into text
Drop a video here and Extractune writes down what's said or sung in it: an interview, a lecture, a voice note, a reel.
- No sign-up
- Audio sent only when you tap Get the words
No file open
Video stays here. Audio sent only on tap.
Drop a file here
Audio or video, up to 500MB, 20 min
Video to text and subtitles
Writes down what's said.
You get timed lines you can search, tap to replay and copy, plus subtitle files, in a few seconds.
How it works
- 01
Drop the video here
Drop it on the tool at the top of this page, or tap Choose a file: any MP4, MOV, WebM or MKV, up to 500MB and 20 minutes. Its sound comes out right in your browser.
- 02
Tap Get the words
What's said is written down in a few seconds, as timed lines under the video.
- 03
Copy it or save subtitles
Copy the text with one tap, or download it as subtitles (SRT or VTT). Only need part of a long video? Trim it on the Convert tab first.
The language is detected for you. Clear speech transcribes best; loud music or crosstalk makes it harder.
What people use it for
- Get the quotes from an interview you recorded.
- Turn a lecture into notes, and find the part you need by searching.
- Caption a reel or a short.
- Read a voice note instead of listening to it.
- Write down the lyrics of a song you recorded.
What you get, and how accurate it is
The transcript comes back as timed lines, not one block of text. The line being said lights up as the video plays, tapping a line plays the video from there, and a search box finds every line with a word in it, so a 20-minute lecture is quick to scan. The same lines show on the video as captions.
It's written by Whisper large-v3-turbo, a speech model trained on many languages. The language is found from the audio itself, so a video in Hindi, Spanish or Tagalog needs no setting. It writes down the language that's spoken; it doesn't translate.
Clear speech close to the microphone comes out best. Loud background music, people talking over each other or heavy echo make it less accurate, and it doesn't label who is speaking. When the video is a known song, a Lyrics tab appears beside the transcript too.
Only the sound leaves your device, compressed to a small mono file; the video never does. Clips can be up to 20 minutes, with about 10 transcripts an hour for each visitor, which keeps it free.
Questions
How do I transcribe a video?
Drop the video on this page and tap Get the words. The transcript appears in a few seconds, ready to copy.
Which languages work?
Most widely spoken languages. The language is detected from the audio itself, so there's nothing to choose.
Can I get subtitles instead of plain text?
Yes. Tap Subtitles above the transcript to download it as an SRT or VTT file, timed to the video. If you trim the video in the tool first, you can time them to the whole video or to your cut.
Does the transcript have timestamps?
Yes. Every line starts with the time it's said, in the video's own time, and tapping it plays from there. Copy gives you the plain text; the subtitle files keep the timings.
Can it translate the video?
No. It writes down the words in the language they're spoken in. Copy the text into a translator if you need it in another language.
Can I transcribe an audio file, like a voice memo?
Yes. Drop an MP3, M4A, WAV or other audio file here, the same way as a video.
Does my video get uploaded?
No, the video never leaves your device. When you tap Get the words, the extracted audio (compressed, never the video) is sent to be transcribed, and a 20-second clip to check whether it's a known song. Extractune doesn't keep either after the request.
Is it free?
Yes, with a fair-use limit on how many videos you can transcribe per hour, to keep it free for everyone.
More ways to use Extractune
Updated