Audio version — Estimated duration: 3 min 15 sec

Installation

Requirements

• Python 3.8+
• CUDA-compatible NVIDIA GPU (tested on RTX 3050 6GB)
• Wayland compositor (GNOME, KDE, Sway, etc.)

System Dependencies

# Ubuntu/Debian
sudo apt install wl-clipboard libnotify-bin libportaudio2
 
# Arch
sudo pacman -S wl-clipboard libnotify portaudio

Python Dependencies

pip install evdev faster-whisper sounddevice numpy

Step-by-Step Setup

1. Clone the repository:

git clone https://github.com/prathmeshnik/Whisper-Dictate.git
cd Whisper-Dictate

2. Create a virtual environment (recommended):

python3 -m venv .venv
source .venv/bin/activate

3. Install Python dependencies:

pip install evdev faster-whisper sounddevice numpy

4. Add your user to the input group (required for keyboard access):

sudo usermod -aG input $USER

Log out and back in (or reboot) for the group change to take effect.

5. Make the script executable:

chmod +x whisper-dictate-v2.py

Auto-Start (Optional)

Systemd Service

mkdir -p ~/.local/bin
cp whisper-dictate-v2.py ~/.local/bin/
chmod +x ~/.local/bin/whisper-dictate-v2.py
cp whisper-dictate.service ~/.config/systemd/user/
systemctl --user daemon-reload
systemctl --user enable whisper-dictate
systemctl --user start whisper-dictate

Desktop Autostart

cp whisper-dictate.desktop ~/.config/autostart/

Next Steps

• Quickstart — Run the script for the first time
• Configuration — Customize settings
• Troubleshooting — If something goes wrong