Menu
AI Captions

Add Subtitles to Video

Word-accurate AI captions in minutes — preview them live on your video, download SRT/VTT for any platform, or burn them in permanently.

AI captions run on our secure servers.

The audio track is extracted in your browser — the video itself never uploads for transcription. Whisper AI generates word-accurate captions (1 credit per minute of audio), exportable as SRT/VTT or burned into the video. A free account includes 6 AI audio credits per month — no card required.

How it works

Step 1

Load your video

The audio track is extracted right in your browser — the video itself stays on your device.

Step 2

Generate captions

Whisper AI transcribes with word-level timing and the captions appear live on your video.

Step 3

Export your way

Download SRT/VTT sidecars, or burn the captions permanently into an MP4.

Frequently asked questions

How accurate are the AI captions?

Captions are generated by Whisper, the state-of-the-art open speech model, with word-level timestamps grouped into readable lines. Clear speech transcribes near-perfectly; heavy accents, crosstalk, or loud music reduce accuracy. Language is auto-detected across 95+ languages.

Does my whole video get uploaded?

Not for transcription — the audio track is extracted in your browser and only that small audio file is sent for AI processing (priced at 1 credit per minute of audio). The full video only uploads if you choose the optional burn-in step.

SRT sidecar or burned-in — which should I use?

SRT/VTT sidecar files keep captions toggleable and are accepted by YouTube, LinkedIn, and every major platform — use them when you can. Burn-in renders the text permanently into the picture, which is right for Instagram, TikTok, or anywhere sidecars aren't supported.

Can I edit the captions before burning them in?

Download the SRT, edit it in any text editor, and keep it as a sidecar. In-app caption editing is on our roadmap — tell us via the Feedback button if you want it prioritized.

Working with audio instead? Transcribe audio to text — or trim your video first.