WaveFit: An Iterative and Non-autoregressive Neural Vocoder based on Fixed-Point Iteration

10/03/2022
by   Yuma Koizumi, et al.
0

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 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 <google.github.io/df-conformer/wavefit/>.

READ FULL TEXT
research
08/28/2023

Voice Conversion with Denoising Diffusion Probabilistic GAN Models

Voice conversion is a method that allows for the transformation of speak...
research
01/07/2021

Knowledge Distillation in Iterative Generative Models for Improved Sampling Speed

Iterative generative models, such as noise conditional score networks an...
research
07/31/2023

DiffProsody: Diffusion-based Latent Prosody Generation for Expressive Speech Synthesis with Prosody Conditional Adversarial Training

Expressive text-to-speech systems have undergone significant advancement...
research
07/01/2019

Analysis by Adversarial Synthesis -- A Novel Approach for Speech Vocoding

Classical parametric speech coding techniques provide a compact represen...
research
11/25/2021

Joint inference and input optimization in equilibrium networks

Many tasks in deep learning involve optimizing over the inputs to a netw...
research
04/17/2021

Multi-Metric Optimization using Generative Adversarial Networks for Near-End Speech Intelligibility Enhancement

The intelligibility of speech severely degrades in the presence of envir...
research
04/22/2023

Conditional Denoising Diffusion for Sequential Recommendation

Generative models have attracted significant interest due to their abili...

Please sign up or login with your details

Forgot password? Click here to reset