Skip to main content

Auto Subtitle Generator — Free AI Captions

Whisper AI writes your captions on your own device.

  • AI runs on your device
  • Never uploaded
  • No watermark
  • No sign-up

Drop a video or audio file here, or click to choose

MP4, MOV, MKV, WebM, AVI, WMV, FLV and more · free up to 15 minutes

AI runs on your device — your file is never uploaded

What this tool does

Drop in a video and VideoMPEG listens to the speech and writes time-coded subtitles for you. It uses Whisper, the open-source speech model from OpenAI, running directly in your browser — so the captions are made on your device and your video is never uploaded anywhere.

When it finishes you can correct any word, switch between full subtitle lines and short punchy phrases, then download an SRT or WebVTT file for YouTube, Vimeo, Premiere Pro, DaVinci Resolve or any player — or burn the captions straight into the picture.

How to use auto subtitle generator

  1. 1

    Choose a video or audio file, or drop it onto the box above.

  2. 2

    Pick the spoken language (or leave it on automatic) and the caption length.

  3. 3

    Click Generate subtitles. The AI model downloads once, then runs on your device.

  4. 4

    Fix any words in the caption editor, then download SRT or VTT, or burn the subtitles into the video.

Tips for better results

Getting the most accurate captions

Clear speech with little background music works best. Choosing the spoken language instead of automatic detection avoids mistakes in the first few seconds, and names are worth a quick check in the editor.

SRT or VTT?

SRT is accepted by YouTube, Facebook, LinkedIn and every editing app. VTT is the web standard used by HTML5 players and sites like Vimeo. Both contain the same captions.

Why your files stay private

Most online converters upload your video to their servers, process it there, and ask you to download the result. VideoMPEG works the other way round: the processing code is sent to your browser, and your file never leaves your device. That’s safer for personal footage, faster for big files, and the reason there are no queues, caps or watermarks.

How VideoMPEG works →

Frequently asked questions

Is my video uploaded to an AI service?

No. The speech model (OpenAI’s open-source Whisper) is downloaded to your browser once and runs on your own device, using your graphics card through WebGPU where available. Your audio and video never leave your computer or phone.

Is it free?

Yes. Files up to 15 minutes are free with no sign-up and no watermark. VideoMPEG Pro removes the length limit and adds a more accurate model and translation to English.

Which languages does it understand?

Whisper recognizes speech in dozens of languages, including English, Spanish, Portuguese, French, German, Italian, Hindi, Japanese, Korean and Chinese. Choose the spoken language for the best results, or let it detect the language automatically.

How do I add the SRT file to YouTube?

In YouTube Studio, open the video, choose Subtitles, add a language, then Upload file and select “With timing”. Pick the .srt file you downloaded here.

Why is the first run slower?

The first time, your browser downloads the speech model (roughly 80–200 MB depending on your device). After that it is cached, so later videos start in seconds.

Are my files uploaded anywhere?

No. VideoMPEG processes files entirely inside your browser using WebCodecs and FFmpeg compiled to WebAssembly. Your video is never sent to our servers or anyone else’s, which is also why there are no upload waits or size caps.

Is it really free?

Yes — no sign-up, no watermark and no daily limits on any converter or editing tool. VideoMPEG is supported by ads, clearly labeled partner links and optional Pro memberships, which add batch processing, unlimited AI subtitles and an ad-free site.