Parallel WaveNet: Fast High-Fidelity Speech Synthesis
This enables real-time, high-quality speech synthesis for applications like virtual assistants, addressing a deployment bottleneck.
The paper tackled the slow sequential generation of WaveNet for speech synthesis by introducing Probability Density Distillation to train a parallel feed-forward network, achieving high-fidelity speech generation at over 20 times faster than real-time and deployment in Google Assistant.
The recently-developed WaveNet architecture is the current state of the art in realistic speech synthesis, consistently rated as more natural sounding for many different languages than any previous system. However, because WaveNet relies on sequential generation of one audio sample at a time, it is poorly suited to today's massively parallel computers, and therefore hard to deploy in a real-time production setting. This paper introduces Probability Density Distillation, a new method for training a parallel feed-forward network from a trained WaveNet with no significant difference in quality. The resulting system is capable of generating high-fidelity speech samples at more than 20 times faster than real-time, and is deployed online by Google Assistant, including serving multiple English and Japanese voices.