The app provides instant feedback on speaking metrics such as Words Per Minute (WPM), filler word usage, and overall pacing using OpenAI's speech recognition models for accurate transcription.

• Real-time Transcription: Converts speech to text on the fly.
• Speech Analysis:
  • WPM (Words Per Minute): Tracks speaking speed in real-time.
  • Filler Word Detection: Identifies and highlights common filler words (um, uh, like, you know, etc.).
  • Pacing Analysis: Visualizes speaking pace trends over time.
• Visual Dashboard: Interactive charts and metrics to visualize performance.
• Mock Mode: Test the application's UI and analysis features without using a microphone.

• Framework: Next.js 16 (App Router)
• Language: TypeScript
• AI Integration: Vercel AI SDK
• Styling: Tailwind CSS 4
• Components: ShadCn
• Visualizations: Recharts
• Animations: Motion
• Testing: Vitest & React Testing Library

• Node.js (v18 or higher recommended)
• pnpm (v9 or higher)
• An OpenAI API Key

1. Clone the repository:

   git clone git@github.com:suveshmoza/SpeechScore.git cd speech-score

2. Install dependencies:

   pnpm install

3. Set up environment variables: Create a .env.local file in the root directory and add your OpenAI API key:

   OPENAI_API_KEY=your_openai_api_key_here

4. Start the development server:

   pnpm dev

5. Open http://localhost:3000 in your browser.

1. Grant Microphone Access: Allow the browser to access your microphone when prompted.
2. Start Recording: Click the microphone button to begin analyzing your speech.
3. View Metrics: Watch as the dashboard updates in real-time with your transcript, WPM chart, and filler word counts.
4. Review: Stop recording to see a summary of your session.

• pnpm dev: Starts the development server.
• pnpm build: Builds the application for production.
• pnpm start: Runs the built production application.
• pnpm test: Runs the test suite.
• pnpm lint: Runs ESLint to check for code quality issues.