Transform your stream-of-consciousness audio recordings into structured, actionable insights.

Funnel is a SwiftUI app that captures spontaneous audio recordings and transforms them into multiple visual representations. Perfect for capturing ideas on-the-go when inspiration strikes.

 🎤 Record → 🤖 Process → 📋 Visualize │ │ │ │ │ ├── Bullet Points │ │ ├── Napkin Sketch │ │ └── Full Transcript │ │ │ └── AI-powered transcription & summarization │ └── One-tap recording with AirPods support 

• Simple one-tap recording with real-time waveform visualization
• Background recording capability
• AirPods/headphone support
• Minimum duration check to ensure quality

• Automatic transcription via OpenAI Whisper
• Smart summarization using Anthropic Claude
• Visual diagram generation
• Fast processing with status updates

• Gradient backgrounds with vibrant colors
• Glassmorphic effects for modern aesthetics
• Smooth animations and transitions
• Dark mode optimized

• Chronological list of all recordings
• Search across transcripts and summaries
• Export options (coming soon)
• Archive and delete functionality

• SwiftUI - Modern declarative UI
• SwiftData - Local persistence
• AVAudioRecorder - Audio capture
• Combine - Reactive programming

• Deno - Modern TypeScript runtime
• OpenAI Whisper API - Audio transcription
• Anthropic Claude API - Content analysis
• RESTful API - Clean communication

• Xcode 15+
• iOS 17+
• Deno 1.40+
• API Keys for OpenAI and Anthropic

1. Clone the repository

   git clone https://github.com/happybits/funnel.git cd funnel

2. Set up the iOS app

   make install # Install swiftformat make build # Build the app

3. Configure the backend

   cd server cp .env.example .env # Add your API keys to .env

4. Start the server

   deno task dev

5. Run the app

   make run # Or open in Xcode and run

make help # Show all available commands make build # Build the app make run # Run on simulator make format # Format Swift code make test # Run tests make clean # Clean build artifacts

funnel/ ├── Funnel/ # iOS app │ ├── Views/ # SwiftUI views │ ├── Models/ # Data models │ ├── Services/ # API and recording services │ └── Utils/ # Helpers and extensions ├── server/ # Deno backend │ ├── routes/ # API endpoints │ ├── services/ # External API integrations │ └── tests/ # Server tests └── docs/ # Documentation 

Upload an audio file for processing.

Request:

• Multipart form data with audio file
• Supported formats: mp3, mp4, wav, m4a

Response:

{ "transcript": "Full transcription text...", "duration": 45.2, "bulletSummary": ["Key point 1", "Key point 2"], "diagram": { "title": "Concept Overview", "description": "Visual representation of the main idea", "content": "ASCII art or mermaid diagram" } }

We welcome contributions! Please see our Contributing Guide for details.

• Follow the SwiftUI Layout Best Practices
• Review the Glassmorphism Implementation Guide for performance tips
• The project uses Xcode's new folder-based structure (files are automatically referenced)
• Run make format before committing
• Write tests for new features

MIT License - see LICENSE for details.

• Design inspiration from modern iOS apps
• Built with ❤️ using SwiftUI and Deno
• Powered by OpenAI and Anthropic APIs