---
title: "Auto Subtitle Generator: SRT, VTT and Burned-In | ToolSugar"
description: "Generate timed SRT and VTT subtitles from any video or audio, free. Speech recognition runs in your browser, then you can burn the captions into an MP4."
canonical: https://toolsugar.com/tools/subtitle-generator
image: https://toolsugar.com/og.png
---

# Auto Subtitle Generator

ToolSugar Auto Subtitles listens to a video or audio file and writes timed captions as SRT, VTT and plain text. Whisper runs in your browser, so the file is never uploaded. Each run counts toward 5 free AI runs in any rolling 24 hours, with no sign up. For video files it can also burn the captions into a new MP4, which is not counted.

Runs on your device

## How it works

1. Drop a video or audio file. It stays on your device, and the first run downloads the speech model.
2. Leave the language on Detect automatically, or pick it yourself if the recording mixes languages.
3. Download the SRT, VTT or text file. For a video, you can then burn the captions into a new MP4.

## Key facts

- **Outputs**: SRT, VTT, plain text, and MP4 with burned-in captions for video files
- **Speech model**: Whisper base, run in your browser. One download of about 77 MB, or 206 MB with WebGPU, then cached
- **Free runs**: 5 in any rolling 24 hours, no sign up. Burning captions into a video is not counted
- **Burn-in fonts**: Noto Sans and Noto Sans Devanagari, which cover Latin, Greek, Cyrillic and Devanagari text

## Common uses

- Adding a subtitle file to a YouTube upload so viewers can turn captions on and off.
- Making a muted social clip readable by burning the words into the picture.
- Giving a training video a VTT track for an HTML5 player or a course platform.

## Questions

### Is my recording uploaded?

No. Whisper runs in your browser, so the audio and picture stay on your device. The run is counted with a hash of browser traits and a hash of your IP, not your file. Speed depends on your device, and noisy audio produces more mistakes, so read the result before you publish.

### Can I burn captions in any language?

Burned-in video works for text in Latin, Greek, Cyrillic and Devanagari script, because those are the fonts we ship. For Japanese, Arabic, Korean and other scripts the tool tells you it cannot draw the text. The SRT and VTT files still work in those languages.

### Why does burning captions take longer than generating them?

Burning re-encodes every frame of the video on your own device, on one processor thread. A short 360p clip finished in about real time in our test. Expect a 1080p video to take much longer than its own length.

[Read the Auto Subtitles guide](https://toolsugar.com/blog/how-to-add-subtitles-to-a-video)

## Related tools

- [Captions for Reels and Shorts](https://toolsugar.com/tools/auto-captions-for-reels)
- [Transcribe to Text](https://toolsugar.com/tools/transcribe-audio-to-text)
- [Video Translator](https://toolsugar.com/tools/translate-video)
- [Trim Video](https://toolsugar.com/tools/video-trimmer)