My theory was that the part of my brain responsible for normal speech was still intact, but for some reason had become disconnected from the neural pathways to my vocal cords. (That's consistent with any expert's best guess of what's happening with Spasmodic Dysphonia. It's somewhat mysterious.) And so I reasoned that there was some way to remap that connection. All I needed to do was find the type of speaking or context most similar -- but still different enough -- from normal speech that still worked. Once I could speak in that slightly different context, I would continue to close the gap between the different-context speech and normal speech until my neural pathways remapped. Well, that was my theory. But I'm no brain surgeon.
The secret (for him) was rhyming!