I am currently trying to do speech recognition in a small-ish space. I am planning to include 2-3 microphones in different areas of this space in hope of getting better speech recognition.

I was wondering if anyone had ideas or knows about ways to perform speech to text with C# and some library/hardware that will take input from 2 or more microphones. I just felt that if I had more microphones the person can move around the space and still be able to get good speech recog.

I know microphone arrays are useful for this, but I am not very familiar with how the arrays work with software. Do I just get one audio stream that's a mix of all 3 microphones?

Has anyone tried this with the SpeechLib that works with C#?