This Python project is a real time transcription audio speech to text from internal audio computer. The main library used is fasterwhisper with large-v3 model, soundcard for audio capture from internal audio computer, WAVE for export audio and Tkinter for GUI.
pip install -r requirements.txt
Python 3.8 or greater CUDA 12 for GPU acceleration
Tested on windows 10 with Nvidia RTX 4060 8GB