Voice 2 Literacy is an interactive web-based elementary phonics reading room designed for young learners. It allows children to read stories, practice pronunciation via native Web Speech synthesis, dictate stories using Speech Recognition, and receive real-time, dynamic guidance from Toby the Tutor—an AI agent rendered dynamically using Google's A2UI (Agent-to-User Interface) protocol.
This project was built as a capstone project for Kaggle's 5-Day AI Agents: Intensive Vibe Coding Course with Google.
- Interactive Phonics Stories: Explore sound focus stories (e.g. The Friendly Fox, Sam the Fat Cat, Ben's Big Red Bus). Click on any word to hear it read aloud with kids-oriented speed and pitch.
- Sound Hunt Game: Complete challenges by finding and clicking words containing the targeted phonics sounds. Highlights progress and updates in real-time.
- Expressive Dictation Canvas: A child-friendly canvas where kids can type or click "Voice Type" to transcribe speech to text. Offers real-time phonics milestones tracking how many focus sound words they've typed.
- Toby the AI Tutor (A2UI Surface): Powered by a simulated AI agent session using
@a2ui/litand@a2ui/web_core. Toby dynamically streams visual feedback cards, traces, and sticker medals (like the 🏆 Sound Master Badge) directly to the A2UI surface in response to user actions. - Developer Console: Toggle open the built-in debugger to inspect the exact A2UI JSON message streams (e.g.
createSurface,updateComponents) being processed by the renderer.
- Frontend Framework: Lit (Web Components) + TypeScript
- State & Rendering Engine: Google A2UI (
@a2ui/lit,@a2ui/web_core) - Build Tool: Vite
- Speech APIs: Web Speech API (
SpeechSynthesis,webkitSpeechRecognition) - Styling: Child-friendly design tokens (using Fredoka and Outfit Google Fonts)
- Node.js (v18+)
- npm (v10+)
- Clone this repository to your local machine.
- Navigate to the directory:
cd voice-2-literacy - Install dependencies:
npm install
- Start the Vite local development server:
npm run dev
- Open your browser and navigate to the local server URL (usually
http://localhost:5173/).
This project serves as my capstone project, designed and developed during the Kaggle & Google AI Agents Intensive (June 15 – 19, 2026).
Watch the Capstone Demo Video on YouTube
- Laquita Jordan - Lead Developer
- Cohort: Kaggle & Google AI Agents Intensive (June 2026)
- 📧 Email: laquitaj901@gmail.com
- 💼 LinkedIn: laquitajordan-memphis
- 🌐 Portfolio Website: MyDataFolio Profile