In Support of the Design of Intelligent Speech Prostheses