Audio Chat App

A voice-first conversation application built with Tauri and React that enables natural audio conversations with AI. This app provides a seamless push-to-talk experience for macOS users.

🎯 Features

• Voice-First Interface: Push-to-talk audio recording with real-time waveform visualization
• Speech-to-Text: Powered by OpenAI's Whisper model for accurate transcription
• AI Conversations: Integration with Ollama for intelligent responses using Gemma3N model
• Text-to-Speech: macOS native TTS for natural voice responses
• Conversation History: Persistent storage of all conversations and messages
• Modern UI: Beautiful gradient interface with responsive design
• Permission Management: Proper microphone permission handling for macOS

🛠️ Tech Stack

Frontend

• React 19 with TypeScript
• Tailwind CSS for styling
• Vite for build tooling
• Tauri API for native integration

Backend

• Rust with Tauri 2.0
• Whisper-rs for speech recognition
• Ollama for AI chat (Gemma3N model)
• SQLite for data persistence
• cpal for audio processing
• macOS native TTS for speech synthesis

📋 Prerequisites

Before running this application, ensure you have:

1. Node.js (v18 or higher)
2. pnpm package manager
3. Rust toolchain (latest stable)
4. Ollama installed and running
5. macOS (for native TTS and permissions)

🚀 Installation

1. Clone the Repository

git clone https://github.com/Dhruv2mars/pR-t-04.git
cd pR-t-04/audio-chat-app

2. Install Dependencies

pnpm install

3. Set Up Ollama

Install Ollama and run the required model:

# Install Ollama (if not already installed)
curl -fsSL https://ollama.ai/install.sh | sh

# Pull and run the Gemma3N model
ollama pull gemma3n:latest
ollama run gemma3n:latest

4. Download Whisper Model

Download the Whisper model for speech recognition:

# Create models directory
mkdir -p src-tauri/models

# Download the Whisper model (you'll need to download this manually)
# Place ggml-base.en.bin in src-tauri/models/

5. Build and Run

# Development mode
pnpm dev

# Build for production
pnpm build

🎤 Usage

1. Launch the App: Start the application and grant microphone permissions when prompted
2. Start Conversation: Click and hold the microphone button to record your message
3. Release to Send: Release the button to send your audio message
4. Listen to Response: The AI will respond with synthesized speech
5. View History: Access conversation history from the sidebar

📁 Project Structure

audio-chat-app/
├── src/                    # React frontend
│   ├── components/         # React components
│   │   ├── VoiceChat.tsx  # Main voice interface
│   │   ├── HistorySidebar.tsx
│   │   ├── StatusBar.tsx
│   │   ├── Waveform.tsx
│   │   └── MicrophonePermissionModal.tsx
│   ├── App.tsx            # Main app component
│   └── main.tsx           # Entry point
├── src-tauri/             # Rust backend
│   ├── src/
│   │   ├── main.rs        # Tauri app setup
│   │   ├── asr.rs         # Speech recognition
│   │   ├── tts.rs         # Text-to-speech
│   │   ├── ollama.rs      # AI chat integration
│   │   ├── db.rs          # Database operations
│   │   ├── microphone.rs  # Microphone handling
│   │   └── audio_processing.rs
│   └── Cargo.toml         # Rust dependencies
└── package.json           # Node.js dependencies

🔧 Configuration

Environment Variables

The app uses default configurations, but you can customize:

• Ollama URL: Defaults to http://localhost:11434
• Whisper Model: Uses ggml-base.en.bin
• TTS: Uses macOS native say command

Database

Conversations are stored in SQLite database located in the app's data directory.

🐛 Troubleshooting

Microphone Permissions

If microphone access is denied:

1. Go to System Preferences > Security & Privacy > Privacy > Microphone
2. Add the app to the list of allowed applications
3. Restart the application

Ollama Connection Issues

• Ensure Ollama is running: ollama run gemma3n:latest
• Check if port 11434 is accessible
• Verify the model is downloaded: ollama list

Whisper Model Issues

• Ensure ggml-base.en.bin is in src-tauri/models/
• Check file permissions and size
• Verify the model file is not corrupted

Audio Processing Issues

• Check microphone input levels
• Ensure recording duration is at least 0.5 seconds
• Verify audio format compatibility

🤝 Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feature-name
3. Make your changes
4. Test thoroughly
5. Submit a pull request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• OpenAI for the Whisper speech recognition model
• Ollama for the local AI inference
• Tauri for the cross-platform framework
• React and Rust communities

📞 Support

For issues and questions:

• Create an issue on GitHub
• Check the troubleshooting section above
• Ensure all prerequisites are met

---

Note: This application requires macOS for optimal functionality due to native TTS and permission handling.