https://github.com/namuan/meeting-buddy
Making meetings more productive, one transcription at a time! 🚀
https://github.com/namuan/meeting-buddy
coach meeting-notes ollama transcriber whisper
Last synced: about 1 month ago
JSON representation
Making meetings more productive, one transcription at a time! 🚀
- Host: GitHub
- URL: https://github.com/namuan/meeting-buddy
- Owner: namuan
- License: mit
- Created: 2025-08-25T07:50:37.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-09-14T20:03:17.000Z (9 months ago)
- Last Synced: 2025-09-14T21:34:34.741Z (9 months ago)
- Topics: coach, meeting-notes, ollama, transcriber, whisper
- Language: Python
- Homepage:
- Size: 4.9 MB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Meeting Buddy 🎤
A real-time audio recording and transcription application with AI-powered analysis capabilities. Meeting Buddy helps you capture, transcribe, and analyze meetings or conversations using OpenAI Whisper for speech-to-text and local LLM integration via Ollama.
## Features
- **Real-time Audio Recording**: Capture audio from any input device
- **Live Transcription**: Convert speech to text using OpenAI Whisper
- **AI Analysis**: Process transcriptions with local LLMs via Ollama
## Requirements
- Python 3.10 (specifically, not 3.11+)
- PyQt6 for the GUI
- PyAudio for audio recording
- OpenAI Whisper for transcription
- Ollama (optional, for LLM analysis)
## Installation
### Prerequisites
1. **Install Python 3.10**: Make sure you have Python 3.10 installed (not 3.11+)
2. **Install Ollama** (optional): For LLM analysis features
```bash
# macOS
brew install ollama
# Or download from https://ollama.ai
```
### Quick Start
1. **Clone the repository**:
```bash
git clone https://github.com/namuan/meeting-buddy.git
cd meeting-buddy
```
2. **Install dependencies**:
```bash
make install
```
3. **Run the application**:
```bash
make run
```
## Usage
### Basic Usage
1. **Launch the application**:
```bash
make run
```
2. **Select Audio Input**: Choose your preferred audio input device from the dropdown
3. **Start Recording**: Click the "Start Recording" button to begin capturing audio
4. **View Transcription**: Watch as your speech is transcribed in real-time
5. **AI Analysis** (optional): If you have Ollama running, enter a prompt to analyze the transcription
6. **Stop Recording**: Click "Stop Recording" when finished
### Audio Setup (macOS)
Install BlackHole (via brew)
```shell
brew install blackhole-2ch
```
For optimal audio capture on macOS:
1. Open **Audio MIDI Setup** (Applications > Utilities)
2. Configure your audio devices as shown:
> [!WARNING]
> If you can't see BlackHole in the list, you may need to `sudo killall coreaudiod`
3. Set the correct sound output device:
## Configuration
### LLM Integration
To use LLM analysis features:
1. **Install Ollama**: Follow the installation instructions above
2. **Start Ollama**:
```bash
ollama serve
```
3. **Pull a model** (e.g., llama3.2):
```bash
ollama pull llama3.2
```
4. **Configure in app**: The application will automatically detect running Ollama instances
## Troubleshooting
### Common Issues
1. **No audio devices detected**:
- Check that your microphone is connected and recognized by the system
- Try running with verbose logging: `make run -v`
2. **Transcription not working**:
- Ensure OpenAI Whisper is properly installed
- Check audio input levels
- Verify the selected audio device is working
3. **LLM analysis not available**:
- Make sure Ollama is installed and running
- Verify you have pulled at least one model
- Check Ollama is accessible at the default port (11434)
4. **Application won't start**:
- Ensure you're using Python 3.10 (not 3.11+)
- Try reinstalling dependencies: `make install`
- Check the logs for specific error messages
### Logging
For debugging issues, run with increased verbosity:
```bash
uv run meeting-buddy -vv
```
Logs will help identify issues with audio devices, transcription, or LLM connectivity.
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.