1- Paste your openAI secret key in gpt_key.txt file. You can get your key from here : key
2- Install Pytorch here (choose cuda if you have a good GPU else choose CPU)
3- Install chocolatey here (for windows)
4- Install ffmpeg "choco install ffmpeg" in powershell as admin (for windows)
5- Install python requirements "pip install -r requirements.txt"
After installing all required modules run the script hold "v" key wait for 5 secs while holding and then you can start talking. When you stop talking wait for a couple of secs and then release the "v" key this way you will ensure the recogniser captures your whole voice.