At a glance
- Accepts
- MP4, WebM, MOV, MP3, WAV, M4A and other files the browser can decode, up to 60 minutes (20 on phones)
- Outputs
- An editable list of captions, previewed over the video, downloaded as .srt or .vtt
- Where it runs
- Entirely in your browser, on your device
- What leaves your device
- Nothing - the file never leaves your device
- Limits
- One timestamp per phrase from Whisper, so times are estimated where a phrase is split; the language must be picked, not detected; the whole audio is held in memory; downloads a speech model once; no burn-in
- Account
- Not required - there is no sign-up
- Price
- Free, with no watermark and no usage cap
- Made by
- Visrut
- Last updated
- 11 November 2026
Subtitle Generator listens to a video or audio file with the Whisper speech model and writes timed captions you can download as .srt or .vtt. Captions are cut to a readable length, shown over your video as it plays, and every one can be edited first. It all runs in your browser, so the file stays on your device.
How to use it
- Choose a video or audio file, or drop it here.
- Pick the spoken language and a model, then press Make subtitles.
- Set the characters per line, lines per caption and longest caption time.
- Play the video, fix any wrong words or times, and add or remove captions.
- Download the .srt or .vtt file.
Frequently asked questions
Is my video uploaded anywhere?
No. Your browser reads the sound from the video and the Whisper speech model runs inside the tab. The only download is the model itself, the first time you use it. The video never leaves your device.
What is the difference between SRT and VTT?
Both are plain text files with a start time, an end time and the words for each caption. SRT is the older format and works almost everywhere, including YouTube, Premiere, DaVinci Resolve and VLC. VTT (WebVTT) is the format web video players use with the HTML track element. If you are not sure, download SRT.
How long should a caption line be?
A common guideline is 42 characters per line, two lines at most, and no longer than about 6 seconds on screen. Those are the defaults. For vertical video such as Reels, Shorts or TikTok, try 25 to 30 characters and one line.
How accurate are the captions?
Clear speech in a quiet room comes out well; music, crosstalk, accents and names the model hasn't heard bring mistakes. Whisper gives one time per phrase rather than per word, so when a phrase is split across captions the split times are estimated. Every caption stays editable, with its own play button, before you download.
Can it burn the subtitles into the video?
Not yet. This tool makes the subtitle file, which you add in your video editor or upload alongside the video. Most platforms prefer a separate file anyway, because viewers can turn it off and it can be translated.