WaveFit: An Iterative and Non-autoregressive Neural Vocoder based on Fixed-Point Iteration
This is an incremental improvement for speech synthesis, addressing speed and quality bottlenecks in neural vocoders.
The paper tackled the problem of slow inference in neural vocoders by proposing WaveFit, which integrates adversarial training into an iterative denoising framework, achieving naturalness comparable to human speech with five iterations and inference speed over 240 times faster than WaveRNN.
Denoising diffusion probabilistic models (DDPMs) and generative adversarial networks (GANs) are popular generative models for neural vocoders. The DDPMs and GANs can be characterized by the iterative denoising framework and adversarial training, respectively. This study proposes a fast and high-quality neural vocoder called \textit{WaveFit}, which integrates the essence of GANs into a DDPM-like iterative framework based on fixed-point iteration. WaveFit iteratively denoises an input signal, and trains a deep neural network (DNN) for minimizing an adversarial loss calculated from intermediate outputs at all iterations. Subjective (side-by-side) listening tests showed no statistically significant differences in naturalness between human natural speech and those synthesized by WaveFit with five iterations. Furthermore, the inference speed of WaveFit was more than 240 times faster than WaveRNN. Audio demos are available at \url{google.github.io/df-conformer/wavefit/}.