"A high quality text-to-speech system composed of multiple neural networks."

Orhan Karaali et al. (1998)
a service of Schloss Dagstuhl - Leibniz Center for Informatics