Born from necessity
SceneTell was started by cofounders Sina Bahram and Austin Hicks, who are both blind. The company comes from a direct access need, not a market abstraction: we have videos we need described, and so do many other people.
That lived need shapes the product. SceneTell should be practical, fast enough to use, transparent enough to review, and serious enough for the workflows where video access matters.
The product we are building
SceneTell combines media processing and AI-assisted narration. It transcodes input, maps speech and silence, samples visual changes, extracts on-screen text, builds scene and character context, plans narration windows, synthesizes speech, and renders reviewable deliverables.
That structure matters because audio description is not a caption read out loud. It has to say the right thing, in the right amount of time, without fighting the original program audio. The engineering work around timing, rendering, caching, cost control, and operations is part of the product.
What we believe
Audio description should not be reserved only for major studio releases, large production budgets, or platforms that made the right decision early. Video is everywhere. Description is not.
We are not claiming every generated track is final. We are building a serious path from undescribed video to usable, inspectable description.
Founders
- Sina Bahram
- Austin Hicks
Founder bios are coming in a later content pass.