Questions about auto captions, answered

AutoCaptions is a browser-based auto caption and AI subtitle tool. It turns speech into timed captions, lets you edit and style them, and exports a captioned MP4 or SRT/VTT/TXT file.

Yes. The full caption editor is free, and SRT, VTT and TXT downloads are free on every plan. Free video exports include an AutoCaptions watermark and are capped at 720p and 10-minute videos.

There is no trial timer. Free video exports are watermarked, capped at 720p, and limited to 10-minute videos. The editor and subtitle files are free. Paid plans remove the watermark and raise the video limits.

Yes. SRT, VTT and TXT downloads are free on every plan, including the free plan. We do not put subtitle files behind the paid wall.

No. You can open the editor and caption a video without signing in. A Google account is required only when you buy or manage a subscription, so we can attach the plan to you.

The video is read in your browser and is not stored as a caption project on our servers. The extracted audio is sent to the transcription provider so it can return timed captions. Custom fonts uploaded by paid users are stored in Cloudflare R2.

Accuracy depends on the audio: clean speech with one speaker is usually close; accents, crosstalk, background noise and jargon need more correction. We do not claim a universal accuracy percentage. Always review the transcript before publishing.

The editor exposes the transcription and translation languages available from our providers. Language coverage can change; check the language selector in the editor before relying on a specific language for a client deadline.

Free: 10 minutes. Pro: 30 minutes. Max: 60 minutes. The limit applies to the source video you caption, not just the exported clip.

Free exports are capped at 720p and include a watermark. Pro and Max export at up to HD without a watermark. SRT, VTT and TXT files are available on every plan.

Yes. Set the canvas to 9:16 and keep captions inside the platform safe zones. The TikTok, Instagram Reels and YouTube Shorts pages include platform-specific guidance.

Yes. Generate captions in the editor and download SRT, then upload the file in YouTube Studio under Subtitles. You can also burn captions into the video if you need them visible everywhere.

AI translation tracks are available on Pro and Max. Review the translated captions before publishing; machine translation can miss idioms, names and tone.

Yes. The timeline lets you adjust caption start and end times, merge or split lines, and close gaps. You can also change the text and line breaks before export.

The marketing site works on mobile. The editor is designed for desktop-class browsers because video decoding and MP4 export are heavy. For best results, use a recent Chrome, Edge or Safari on a computer.

Recent Chrome, Edge and Safari support the browser video export path. Firefox support depends on WebCodecs availability. If export is unavailable, you can still download SRT/VTT and caption the video in another editor.

Open Account → Subscription and choose “Cancel at period end”, or use the Paddle customer portal link in your receipt. Access continues until the end of the current billing period.

Yes, under the 14-day refund window described in the Refund Policy. Contact Paddle support from your receipt or email us with the order details.

Captions can improve accessibility and watch time, and a transcript gives search engines and AI assistants more text to understand. There is no guaranteed ranking boost from captions alone; the video, title, description and page still matter.

Unreviewed machine captions are a starting point, not a compliance guarantee. WCAG 1.2.2 requires accurate, synchronized captions; review wording, timing, speaker labels and punctuation, especially for public-sector, education or regulated content.

You can caption videos you have the right to process, including client work where you have permission. Do not upload footage you are not allowed to share with a transcription provider. See the Terms for the full license and acceptable-use rules.

We do not train models on your audio. Audio is sent to the transcription provider to return captions; the provider’s own retention and training terms also apply. See the Privacy Policy and Security page for the current data flow.