Whisper server setup

The ideal and most performant configuration for running the OpenAI Whisper sample is with Windows with WSL 2 and an NVidia GPU or a Linux desktop system with an NVidia GPU.

You can also run the Whisper transcriber server on Windows, macOS, or Linux (tested on Ubuntu) systems without an NVidia GPU. The performance will be slower but more than acceptable when using smaller Whisper models.