Auto Subtitle Generator: Free, In Your Browser

Our auto subtitle generator runs in a browser tab and gives you two ways in. Paste what's said and tap the spacebar in time with the audio, which costs nothing beyond the normal tool charge, or hand it to Whisper and get word-level timing back for 5 tokens per started minute. Either way the subtitles animate word by word, in one of 82 styles, and burn into the frame on export. The rendering happens on your machine, so a 4K file doesn't sit in an upload queue for ten minutes first.

The free path: paste, split, tap

Drop a video in, type or paste the script into the captions box, and hit Create captions. If you paste a paragraph it gets chopped into short TikTok-length chunks automatically, four words a line by default, which you can change. Then press Tap to sync. The video plays from zero and every tap (spacebar works, so does Enter) stamps the start of the next line and the end of the previous one.

It feels crude written down. In practice it's fast and weirdly satisfying, and for a 45-second Short you're done in about 45 seconds. It's also exact in a way transcription sometimes isn't, because you already know what the words are. You're only supplying rhythm.

The AI path, and what it actually costs

Auto captions run Whisper with word-level timestamps. Only the extracted audio uploads, not your video, and the charge is 5 tokens for every started minute of audio. A three-minute clip is 15 tokens. For reference, a free account gets 100 tokens a month and anonymous visitors get 4, which is enough to try the tool without an account at all. If you like it, a free account is the cheapest way to keep going.

Songs get their own mode. Plain Whisper skips whole passages of a music mix, because a wall of instrumentation buries the vocal, so Auto lyrics isolates the vocal first and transcribes the clean stem. That's two models instead of one, which is why it's 15 tokens a minute rather than 5. If you already have the lyric sheet, paste it in and the tool cross-checks its own transcription against your text before you export, which catches the mondegreens.

The styles are the actual reason to use it

There are 82 presets and each one animates live in the picker, so you're picking from motion rather than from a static thumbnail. Beast Mode, Hormozi, karaoke fills, neon signs, horror, pixel, handwritten. Filter by vibe with the chips along the top. Every preset is a starting point, and the Customize panel opens up font, size, position, caption width, colors, outline, and the highlight treatment.

You can also drag the words anywhere in the frame. Not a position dropdown with nine anchor points, an actual drag. Useful when the speaker's face is bottom-centre and the safe zone isn't where the template assumed.

Remove ums and uhs while you're in there

There's a filler cleanup button in both caption states, one for the pasted draft and one for the finished rows. It strips ums, uhs, likes and whatever else you add to the word list. Free, client-side, no tokens. I use it on every talking-head clip and it consistently takes out more than I expect, which is humbling. There's a longer walkthrough in removing filler words from video.

Two honest limitations

• Export is realtime. The tool draws every frame and records it, so a five-minute video takes roughly five minutes to export. That's the price of rendering on your machine instead of a server farm, and it's the thing I'd fix first if I could.
• There's no SRT or VTT sidecar yet. Captions burn into the video. You can download the timed rows as a plain text file (one m:ss.xx to m:ss.xx line per caption) which is fine for reference or hand-editing, but it isn't a subtitle file YouTube will ingest. If burned-in captions are what you wanted, this is a feature. If you needed a sidecar, it isn't there.

Getting started

The tool is embedded on its own page and works signed out on the anonymous token grant, so you can caption something right now and decide afterwards. A free account raises that to 100 tokens a month and keeps your projects. For vertical formats specifically, the Shorts captions guide covers safe zones and sizing, and the TikTok captions post goes deeper on style choice.