Speech synthesis has come a long way:
http://www.naturalvoices.att.com/demos/