SafeSteps

• 

  This is the home page of SafeSteps

• 

  These are all the features available in SafeSteps

Inspiration

Watching content creators facing visual impairments and sharing their lives via vlogs on YouTube. We wanted to make this world more accessible for them.

What it does

It offers features like —

1. Scene description: Describes nearby people, objects, and layouts
2. ⁠Navigation and nearest exit: Guides users with step-wise audio cues
3. Text recognition in environment: Reads out all the textual data in your current environment
4. Document reading: Scans printed pages, signs, or screens and reads them aloud
5. ⁠Video description: Adds spoken narration for visual actions in videos ## How we built it Using gemini ## Challenges we ran into Deploying the project, compiling multiple features into a single project. We ran out of api keys during testing. We lost our whole project once and had no backups. ## Accomplishments that we're proud of We created the vision we had in our mind. And the navigation is really description and detailed. ## What we learned It's good to keep backups and keep recording our progress ## What's next for SafeSteps Building an app and voice assistant

Built With

• ai
• alternative
• as
• canvas
• client-side
• database
• gemini
• google
• hosted
• no
• on
• openai
• speech)
• state
• typescript-&-javascript-|-react-(vite)-frontend-|-tailwind-css-for-styling-|-browser-apis-(camera/mediadevices
• vercel
• via
• vision)
• web
• with
• |