Build an Accessibility Checklist for a Voice Health Assistant
Overview
What this challenge is about.
You receive 20 audio samples spanning accents and speech patterns, the assistant's published dialog state machine, and a list of current voice prompts. Audit the assistant for inclusive design and accessibility, focusing on speech-recognition robustness, repair strategies (how the user corrects a misrecognition), confirmation patterns, and time-to-task. Produce: (1) a 25-item accessibility checklist scoped to voice AI, (2) a 30-minute lab evaluation protocol the QA team can rerun each release, (3) a sample scored evaluation of the current build with three concrete redesign recommendations.
The Brief
What you'll do, and what you'll demonstrate.
Make voice-assistant accessibility a repeatable, release-gating check rather than a one-off audit.
Earning criteria — what you'll demonstrate
- Apply WCAG 2.2 and Section 508 to voice-only interfaces
- Design a release-gating evaluation protocol that non-researchers can run
- Identify repair-strategy failures in conversational AI
- Prioritize accessibility findings by user impact, not technical effort
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Product Designer
Owning the accessibility surface of a voice AI product, including a release-gating protocol, is precisely the kind of work AI product designers ship at health and accessibility-sensitive companies.
This challenge sharpens
- accessibility
- voice-interaction-design
- inclusive-design
AI Safety Researcher
Designing reproducible evaluations that catch failure modes before release mirrors the day-to-day of safety researchers focused on real-world harms.
This challenge sharpens
- evaluation
- human-ai-interaction
- inclusive-design
AI Product Manager
Turning audit findings into a prioritized, business-impact-tied recommendation backlog is core AI PM work.
This challenge sharpens
- evaluation
- user-centred-design
- accessibility