I made free and open-source speech-to-text application for Windows. Press key, speak, watch text appear. Both local and cloud models. I took STT app that was local-only, forked it, and made ton of new features, it now can do STT over OpenAI API! And, of course, LLM post processing. This app is Windows only and is FOSS. I think this is most sophisticated speech-to-text app there is. It's all about settings. You can configure everything in this application. https://github.com/MaxITService/AIVORelay