A browser interface based on the Gradio library for OpenAI's Whisper model.

/?__theme=dark to the end of the URL.You can run this on your CPU but it is a lot slower than running it on an GPU.
See this for the GPU hardware requirements.
The models tiny, base and small should run on any hardware.
git clone ...........pip3 install -r requirements.txt or pip install -r requirements.txt.env.example file and rename the copied file to .env.env filepy app.pybash
# Debian-based:
sudo apt update && sudo apt install git python3 python3-venv python3-pip ffmpeg
# Red Hat-based:
sudo dnf install git python3 python3-pip ffmpeg
# Arch-based:
sudo pacman -S git python3 python-pip ffmpeggit clone ...........pip3 install -r requirements.txt or pip install -r requirements.txt.env.example file and rename the copied file to .env.env filepython3 app.py or python app.pyTo use it with a GPU, install Nvidia's CUDA or AMD's ROCm drivers.
To install PyTorch with the right driver version see here: https://pytorch.org/get-started/locally/
Example: To install PyTorch for CUDA version 11.6: pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116
$ claude mcp add whisper-webui \
-- python -m otcore.mcp_server <graph>