TU2-8: Neural Generative Systems for Speech Compression, Synthesis, and Enhancement |
Session Type: Virtual |
Time: Tuesday, November 3, 13:00 - 14:20 |
Virtual Session: View on Virtual Platform |
Location: Room 8 |
Session Chair: Jan Skoglund, Google |
TU2-8-1: A Study on Conditional Features for a Flow-based Neural Vocoder |
Hyungseob Lim; Yonsei University |
Suhyeon Oh; Yonsei University |
Kyungguen Byun; Yonsei University |
Hong-Goo Kang; Yonsei University |
TU2-8-2: Handling Background Noise in Neural Speech Generation |
Tom Denton; Google |
Alejandro Luebs; Google |
Michael Chinen; Google |
Felicia S. C. Lim; Google |
Andrew Storus; Google |
Yero Yeh; Google |
W. Bastiaan Kleijn; Google, Victoria University of Wellington New Zealand |
Jan Skoglund; Google |
TU2-8-3: WaveNetEQ — Packet Loss Concealment with WaveRNN |
Florian Stimberg; DeepMind |
Alex Narest; Google |
Alessio Bazzica; Google |
Lennart Kolmodin; Google |
Pablo Barrera González; Google |
Olga Sharonova; Google |
Henrik Lundin; Google |
Thomas C. Walters; DeepMind |