Heiga Zen, Google
Abstract: Recent progress in generative modeling has improved the naturalness of synthesized speech significantly. In this talk I will summarize these generative model-based approaches for speech synthesis and describe possible future directions.
