Directly modeling speech waveforms by neural networks for statistical parametric speech synthesis.

Keiichi Tokuda, Heiga Zen

Browse the full ICASSP paper archive.