VoiceGraph 🎤

AI-Powered Voice-to-Workflow Automation Platform

Transform your voice commands into automated workflows with visual graph interface. VoiceGraph uses advanced AI to parse natural language, generate executable workflows, and orchestrate complex automation tasks across multiple services.

✨ Features

🎯 Core Capabilities

• Voice-to-Workflow: Speak your automation needs, get visual workflows
• Visual Workflow Builder: Drag-and-drop interface with React Flow
• Real-time Execution: Live progress tracking with node highlighting
• Background Processing: Async execution with complete history tracking
• Multi-Service Integration: Notion, Tavily, Email, GitHub, and more

🤖 AI-Powered

• Cerebras LLM: Ultra-fast workflow parsing and content processing
• Groq Whisper: High-quality speech-to-text transcription
• Smart Parsing: Natural language understanding for complex workflows

🔧 Technical Features

• Parallel Execution: Optimized workflow processing
• Real-time Streaming: Live execution logs and progress updates
• Workflow History: Complete audit trail of all executions
• Voice Editing: Modify workflows using voice commands

🏗️ Architecture

┌─────────────────┐    ┌─────────────────┐
│   Frontend      │    │   AI Services   │
│   (Next.js)     │◄──►│   (Cerebras)    │
│                 │    │   (Groq)        │
└─────────────────┘    └─────────────────┘
         │                       │
         │                       │
         ▼                       ▼
┌─────────────────┐    ┌─────────────────┐
│   Workflow      │    │   Notion API    │
│   Execution     │    │   Integration   │
│   Engine        │    │   Tavily API    │
└─────────────────┘    └─────────────────┘

Data Flow

1. Voice Input → Groq Whisper → Text
2. Text → Cerebras LLM → Workflow JSON
3. Workflow → React Flow → Visual Graph
4. Execution → Direct API Calls → Service APIs
5. Results → Real-time UI Updates

🚀 Quick Start

Prerequisites

• Node.js 18+
• API Keys (see Environment Variables)

1. Clone Repository

git clone <repository-url>
cd voicegraph

2. Environment Setup

Create .env file in the root directory:

# Cerebras AI (Required for prize eligibility)
CEREBRAS_API_KEY=your_cerebras_api_key_here

# Groq (Speech-to-Text)
GROQ_API_KEY=your_groq_api_key_here

# Notion Integration
NOTION_API_KEY=your_notion_integration_key_here

# Tavily (Web Search)
TAVILY_API_KEY=your_tavily_api_key_here

# SMTP (Email)
SMTP_HOST=smtp.gmail.com
SMTP_PORT=587
SMTP_USER=your_email@gmail.com
SMTP_PASSWORD=your_app_password

# Development
NODE_ENV=development

3. Start Services

# Start Frontend
cd frontend
npm run dev

4. Access Application

• Frontend: http://localhost:3000

🎮 How to Use

1. Create a Workflow

• Voice: Click mic button and speak your automation needs
• Text: Type your requirements in the text input
• Templates: Use pre-built workflow templates

2. Configure Workflow

• Enter required parameters (Notion page IDs, email addresses, etc.)
• Review the generated workflow graph
• Modify nodes if needed

3. Execute Workflow

• Real-time: Click "Run Workflow" for live execution
• Background: Toggle "Background Execution" for async processing
• Monitor: Watch real-time progress and logs

4. View Results

• Execution Logs: Real-time progress updates
• Workflow History: Complete audit trail
• Node Status: Visual indicators for success/failure

🔧 API Endpoints

Core APIs

• POST /api/transcribe - Speech-to-text conversion
• POST /api/parse - Natural language to workflow parsing
• POST /api/execute - Real-time workflow execution
• POST /api/execute-background - Background workflow execution
• POST /api/edit-workflow - Voice-based workflow editing

Health & Monitoring

• GET /api/health - System health check
• GET /api/workflow-history - Execution history

🛠️ Development

Project Structure

voicegraph/
├── frontend/                 # Next.js application
│   ├── app/                 # App router pages
│   ├── components/          # React components
│   ├── lib/                 # Utilities and integrations
│   └── public/              # Static assets
└── README.md               # This file

Key Technologies

• Frontend: Next.js 14, React 18, Tailwind CSS, React Flow
• Backend: Node.js, Express.js, Server-Sent Events
• AI: Cerebras LLM, Groq Whisper
• Integrations: Notion API, Tavily API, SMTP

Development Commands

# Start Frontend
cd frontend && npm run dev

# Test connections
curl http://localhost:3000/api/health

🎯 Workflow Examples

Example 1: Notion to Email

Voice: "Get my Notion meeting notes and email me a summary" Workflow:

1. Fetch Notion page
2. Summarize content with AI
3. Send email with summary

Example 2: Web Research

Voice: "Research hotels in Paris and create a Notion page with the results" Workflow:

1. Search web for hotels
2. Process and format data
3. Create Notion page with results

Example 3: Multi-Step Automation

Voice: "Find trending topics, research them, and send a daily digest" Workflow:

1. Search for trending topics
2. Research each topic
3. Summarize findings
4. Create formatted report
5. Send email digest

🔒 Security

API Key Management

• All API keys stored in environment variables
• No hardcoded credentials in codebase

🐛 Troubleshooting

Common Issues

Workflow Stuck in "Pending"

• Verify API keys are set correctly
• Check console logs for errors

Voice Input Not Working

• Ensure microphone permissions are granted
• Check Groq API key is valid
• Verify network connectivity

Background Execution Issues

• Verify workflow history API is accessible
• Check console logs for detailed errors

Debug Commands

# Test API endpoints
curl http://localhost:3000/api/health

📊 Monitoring

Health Checks

• Frontend: http://localhost:3000/api/health

Logs

• Frontend: Browser console + terminal

🏆 Prize Eligibility

This project is designed for hackathon competition with the following sponsor requirements:

Cerebras AI (Required)

• Primary LLM for workflow parsing
• Content processing and summarization
• Models:qwen-3-32b

Meta Llama (Automatic)

• Using Llama models through Cerebras
• Covered by Cerebras integration

🤝 Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

🙏 Acknowledgments

• Cerebras AI for ultra-fast LLM inference
• Groq for high-quality speech-to-text
• Notion and Tavily for API integrations
• React Flow for workflow visualization

---

VoiceGraph - Transform your voice into powerful automation workflows! 🎤✨