This version has improved end of speech detection from 3s to 0.5s This is essential to build a chatbot in iOS with the SFSpeechRecognizer.
I learned this process architecture in a huge warehouse management system. It was really resilient. I wonder if I can use it in smaller projects.