Funding
Funding not disclosed
Founders
Product
Problem
Video creators and organizations must provide audio description to meet WCAG 2.1 AA, ADA, FCC, and Section 508 standards, but producing high‑quality narration manually is labor‑intensive, costly, and difficult to scale across large media libraries.
Solution
AudiodescriptionAI offers a fully automated, cloud‑native platform that generates compliant audio description from uploaded video files. The service extracts dialogue, performs multimodal scene analysis with Gemini 3 Pro models, and creates frame‑accurate narration scripts. Scripts are synthesized into natural‑sounding voiceovers using Google Cloud Text‑to‑Speech and merged with the original video. Users receive DRM‑protected downloads or streaming URLs, and a built‑in accessible player lets end‑users toggle descriptions or interact via live Gemini voice. The end‑to‑end workflow requires only a video upload, after which the platform delivers compliance‑ready content in minutes.
Target Audience
Primary customers are public entities, media companies, educational institutions, and healthcare organizations that must meet ADA and Section 508 video accessibility requirements.
Features
- Automatic transcription and timing extraction of existing audio tracks
- Multimodal AI scene analysis (Vertex AI Gemini 3 Pro) to generate precise visual context descriptions
- Choice of Standard, Extended, or Hybrid formatting for description length and detail
- High‑quality, natural voice synthesis via Cloud Text‑to‑Speech with studio‑grade narration
- Serverless job architecture (Cloud Run) for massive parallel processing and instant scaling
- Secure portal with encrypted uploads and DRM‑protected download or adaptive‑bitrate streaming delivery
- Integrated accessibility‑focused video player with toggleable audio description and live Gemini voice interaction
- Compliance documentation (Letter of Intent) to support audit readiness for DOJ standards