Using Nvidia NeMo AI toolkit to transcribe audio in real-time and display the transcribed text.
Requires CUDA capable GPU to run on the local machine, if setup using virtual audio cables can transcribe the audio that is being played in real-time without any other requirements.
Nvidia NeMo toolkit is under Apache 2.0 license. Which can be found in it's own repo.
For all the other sources and code;
Copyright (c) 2020 Deniz Ariyan