3 ms·

I had higher expectations of this. I was hoping they would release an Omni type model that could handle voice input and output voice. Oh well.