About Seen Aloud

Seen Aloud is a small, independent service that adds audio description to prerecorded video automatically. It was started in 2026 by Isaacc, who also writes the guides on this site.

Why it exists

Audio description is the part of video accessibility that most teams skip. Captions are cheap and quick; human description usually costs $7 to $15 or more per minute and takes days, so course libraries, training videos and product videos rarely get it. WCAG 2.1 AA still asks for it (success criterion 1.2.5), and in the US the ADA Title II rule makes that a deadline for public universities and local governments. Seen Aloud is built to make description affordable enough to actually do.

How the service works

Everything is automated. Speech recognition finds the pauses, a vision model reads the frames, a language model writes short descriptions for those pauses, and a second pass checks every line against the picture before it is voiced and mixed. You get a described MP4, captions, a WebVTT description track, the script and a compliance note. See the full process.

What it does not do

No person reviews your video or the output, and AI descriptions can contain mistakes. Please review the files before you publish them. Our compliance notes and guides are general information, not legal advice or a certification.

How we handle your videos

Uploaded videos are deleted when processing finishes, generated files are deleted after 30 days, and your videos are never used to train AI models. Payments go through Stripe. Details are in our privacy policy.

Contact

Email [email protected] for support, or [email protected] for invoices. A person reads every message.

Postal address: Seen Aloud, Room 1205, 12/F, Central Plaza, 18 Harbour Road, Wan Chai, Hong Kong.