The fundamental frequencies of voice are below half that, I think that means you can use it to determine what is said writing it's sensitivity range.