Installation

Pre-built Packages

Download from GitHub Releases.

On Linux, choose AppImage (universal), .deb (Debian/Ubuntu with sudo apt install ./voice-notepad_*.deb), or tarball. The package installs to /opt/voice-notepad/ and creates a desktop entry.

On Windows, choose the installer (.exe) or portable .zip.

From Source

Step 1: Install System Dependencies

Install required system packages:

sudo apt install python3 python3-venv ffmpeg portaudio19-dev libc++1

Package purposes:

• ffmpeg - Audio format conversion and compression
• portaudio19-dev - Audio recording library headers
• libc++1 - Required by TEN VAD for voice activity detection

Step 2: Clone and Run

Clone and run:

git clone https://github.com/danielrosehill/Voice-Notepad.git
cd Voice-Notepad
./run.sh

The script creates a virtual environment in app/.venv/, installs Python dependencies, and launches the application.

For manual setup:

cd app
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python -m src.main

Building Packages

To build your own Debian package:

sudo apt install dpkg fakeroot
./build.sh --deb
./build.sh --install

Troubleshooting

If PyAudio fails to install, ensure PortAudio headers are available:

sudo apt install portaudio19-dev

If you encounter audio problems:

pactl info           # Check PipeWire/PulseAudio status
pactl list sources short  # List input devices