Automated video descriptions and Q&A for people who are blind or have low vision
From Human-Powered to Automated Video Description for Blind and Low Vision Users
['FUNDING_R01'] · ARIZONA STATE UNIVERSITY-TEMPE CAMPUS · NIH-11132890
An AI system that creates spoken descriptions of online videos and answers questions to help people who are blind or have low vision understand video content.
Quick facts
| Phase | ['FUNDING_R01'] |
|---|---|
| Study type | Nih_funding |
| Sex | All |
| Sponsor | ARIZONA STATE UNIVERSITY-TEMPE CAMPUS (nih funded) |
| Locations | 1 site (SCOTTSDALE, UNITED STATES) |
| Trial ID | NIH-11132890 on ClinicalTrials.gov |
What this research studies
As someone with vision loss, I rely on audio descriptions to know what happens in videos, but volunteer-created descriptions can take too long to appear. This project aims to train AI to automatically generate clear audio descriptions for online videos and to answer questions I might ask about them. The team will build on the YouDescribe platform, combine human examples with machine learning, and test the system with blind and low-vision users to improve accuracy and usefulness. The work will include real-user feedback so the descriptions and answers match what I need in daily life.
Who could benefit from this research
Good fit: Ideal participants are adults with blindness or significant low vision who regularly use online videos and are willing to try automated audio descriptions and ask questions about video content.
Not a fit: People without reliable internet access, those who do not use online video, or users who need highly specialized domain expertise in videos may not benefit immediately.
Why it matters
Potential benefit: If successful, this could give faster, wider access to the information in online videos so I can independently learn, work, and stay safe in situations that rely on video content.
How similar studies have performed: There are existing AI tools for captioning and some pilot work on automatic descriptions, but fully accurate, user-friendly automated audio descriptions and question-answering for blind users remain novel and under active development.
Where this research is happening
SCOTTSDALE, UNITED STATES
- ARIZONA STATE UNIVERSITY-TEMPE CAMPUS — SCOTTSDALE, UNITED STATES (ACTIVE)
Researchers
- Principal investigator: FAZLI, POOYAN — ARIZONA STATE UNIVERSITY-TEMPE CAMPUS
- Study coordinator: FAZLI, POOYAN
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page below.