Video to text, on your device.
Free, no signup, no watermark. Your video never leaves your device.
A transcript of every word, then SRT or VTT in one tap.
Transcribe my videoNo clip handy? Tap Try a sample in the editor.
How do I convert a video to text for free?
Open the video in Cutmallow's free editor and go to Captions. Whisper, an open-source speech-recognition model, transcribes every word on your own device in any of 99 languages. Fix any word in the transcript, then download it as an SRT or VTT subtitle file, or burn it into the video as animated captions. Nothing is uploaded and there's no per-minute limit.
Updated
- Languages
- 99
- Timing
- Word by word
- Formats
- SRT and VTT
- Model
- 130 MB, once
How to do it in three steps
Drop your video into the editor. It opens on your device.
Open Captions. Every word is transcribed with its timing.
Fix any word, then download SRT or VTT, or export the video with captions.
Then keep going in the same place: add captions, zooms and a style, so the Short is finished before you leave the page.
How it works
Cutmallow runs Whisper base, OpenAI's open-source speech-recognition model, inside your browser. The sound is first split at pauses into short pieces, so long videos never go through the model in one go and silence is not mistaken for words.
Each word comes back with its start and end time. That is what lets the same transcript become an SRT or VTT file for YouTube, Premiere or any player, and word-by-word animated captions in the video itself.
Tips for a better result
- Record with the microphone close to your mouth: clean sound gives the most accurate transcript.
- Cut the pauses first; the transcript and captions follow the cuts.
- Upload the SRT with your YouTube video for searchable closed captions.
One flow instead of three apps
Every free tool here opens the same Cutmallow editor, so the whole Short gets finished in one place.
- Pauses, noise and captions in one editor, instead of a different app for each job.
- Private: the video is processed on your device and never uploaded.
- Works on phones as well as laptops, in the browser you already have.
- No credits, no usage limits, no watermark.
Frequently asked questions
Something else? Write to [email protected].
Can I download the transcript as a file?
Yes. The Captions panel has SRT and VTT downloads. Both are plain text files with timings, which YouTube, video players and editors read as subtitles.
Which languages does it understand?
The 99 languages Whisper was trained on, including English, Spanish, Hindi, Arabic, French, German, Portuguese, Japanese and Chinese. The language is detected for you, or you can pick it.
Why does the first transcription take longer?
The speech model downloads the first time you use Captions, about 130 MB, and is kept on your device afterwards, so later videos start right away.
How long can my video be?
Up to 15 minutes. Most Shorts and Reels are well under that, so you can drop a whole take and keep the best part.
Is my video sent to a transcription server?
No. Your video never leaves your device. The model runs in your browser, so the words never leave your device.