This thing takes an audio stream, makes chunks from it and sends it to the Azure Speech service to transcribe it.
Currently it takes a fixed, hardcoded stream. It is also still missing pieces of transcribed text, so I guess not all the chunks are transcribed. Still need to figure this out.
I used this page a lot to learn how to deploy dockerized python to an Azure Container Registry from Github. https://towardsdatascience.com/a-simple-guide-to-deploying-a-dockerized-python-app-to-azure-29753ee507eb