Parallel WaveNet: Fast High-Fidelity Speech Synthesis

11/28/2017

∙

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.

READ FULL TEXT

Parallel WaveNet: Fast High-Fidelity Speech Synthesis

HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis

WaveFlow: A Compact Flow-based Model for Raw Audio

CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows

StyleMelGAN: An Efficient High-Fidelity Adversarial Vocoder with Temporal Adaptive Normalization

Efficient Neural Audio Synthesis

WOLONet: Wave Outlooker for Efficient and High Fidelity Speech Synthesis

Fast DCTTS: Efficient Deep Convolutional Text-to-Speech

Parallel WaveNet: Fast High-Fidelity Speech Synthesis

Related Research

HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis

WaveFlow: A Compact Flow-based Model for Raw Audio

CaloFlow II: Even Faster and Still Accurate Generation of Calorimeter Showers with Normalizing Flows

StyleMelGAN: An Efficient High-Fidelity Adversarial Vocoder with Temporal Adaptive Normalization

Efficient Neural Audio Synthesis

WOLONet: Wave Outlooker for Efficient and High Fidelity Speech Synthesis

Fast DCTTS: Efficient Deep Convolutional Text-to-Speech