Here is a quote from Apple-version of SAM's documentation (might be in C64 too). Anyway,
"We want to create a system that will synthesize any English utterance. One way to begin would be to record every possible utterance on tape and just play back the right one whenever we need it. This would take up more tape or computer memory than could ever exist..."
I'm pretty sure it has been done (but not with tapes!)... or close enough for all practical purposes. Isn't living in the future awesome? Merry Christmas!