Video Description Generator

The problem

Most video lacks audio description, excluding blind and low-vision viewers from visual context that sighted audiences receive automatically.

My contribution

Full-stack engineer on a hackathon prototype

The approach

Chunked video with FFmpeg, combined speech context with multimodal frame analysis, and synchronized generated narration with playback.

The outcome

Produced a working Llama 4 Hackathon prototype that turns uploaded video into timed, customizable audio description.

What I learned

Accessibility work benefits from treating timing, dialogue, and user control as first-class product constraints rather than post-processing details.

Inside the project