High-order sequence modeling using speaker-dependent recurrent temporal restricted boltzmann machines for voice conversion.

Toru Nakashika, Tetsuya Takiguchi, Yasuo Ariki

Browse the full Interspeech paper archive.