FREE TOOL · ON YOUR DEVICE
Automatic subtitles for videos free
Add a video or audio file with speech and Cocuyo writes the subtitles on your device, without uploading the file. Fix each card, pick one of 5 styles (active word in lime, karaoke, bold outline…) and download the SRT, the VTT or the video with the subtitles already on it.
- Free, no watermark
- SRT, VTT and video
- Your video never leaves your device
- Add your video or audio
- Transcribe and fix
- Choose the style and download
HOW TO USE IT
Three steps.
Add your video or audio
Drop or choose an MP4, WebM, MOV or an audio file with speech. Free transcription accepts up to 3 minutes; choose the language (Spanish by default, English or detect).
Transcribe and fix
Press “Transcribe”. When it finishes, check the 1–3 word cards: change the text, merge, split or move the times while you watch the preview over the video.
Choose the style and download
Classic, box, bold outline, active word or karaoke; position and size. Download the SRT and VTT, or create the video with the subtitles burned in.
NEXT STEP
Keep going with these tools
Auto subtitles: common questions
How accurate is the transcription?
With clear speech and little noise it usually gets most words right, but it can make mistakes with names, background music or several people talking at once. That is why every card can be fixed before you create the video. “High” is more accurate than “Normal”, but downloads more and takes longer.
What is the difference between SRT, VTT and the video with subtitles?
SRT and VTT are text files with the timings: you upload them separately to YouTube, social networks or a player, and they can be turned on or off. The video with subtitles has them drawn on the picture, in the style you choose: they show everywhere, even with the sound off.
How long does it take?
It depends on the device: on a recent computer, about as long as the audio lasts; on a phone it can take longer. Creating the video also takes about as long as it lasts. You can cancel at any time.
Can I use it without transcribing?
Yes. You can import an SRT or VTT you already have, or type the subtitles by hand, and use the editor and styles the same way.
What does the transcription?
Whisper (OpenAI’s speech recognition model, MIT-licensed weights) run by transformers.js (Apache 2.0 licence) inside your browser. The first time the model downloads (about 77 MB in “Normal” and about 250 MB in “High”) and your browser keeps it; the audio is never sent to any server.
Are my videos uploaded to a server?
No. The tool runs inside your browser: videos are read and processed on your device and are not sent to Cocuyo or to any third party. The libraries are served from our own site, with no external services. You can check it in your browser’s network tools: your file is not sent.
Is it really free?
Yes. You can try it before signing up; to download the result we ask for a free account, with no subscription. Your free account comes with 30 welcome uses; after that, each use costs 1 credit (top-ups from US$1), and they are unlimited with Basic or Pro. It never adds a watermark.
What are the limits?
Videos up to 300 MB and 10 minutes that your browser can play (MP4, WebM or MOV), with output up to 1080p. Saving video needs WebCodecs, found in recent versions of Chrome, Edge, Firefox and Safari; if it is missing, we tell you.
Learn how to…
MORE FREE TOOLS
Other video tools
All free tools
WHEN YOU WANT TO CREATE
Make new videos with AI.
Cocuyo’s Studio brings 30+ AI models for image, video and voice together. You see the credit cost before generating, and credits never expire.