A Flow-Based Neural Network for Time Domain Speech Enhancement

Abstract

Speech enhancement involves the distinction of a target speech signal from an intrusive background. Although generative approaches using Variational Autoencoders or Generative Adversarial Networks (GANs) have increasingly been used in recent years, normalizing flow (NF) based systems are still scarse, despite their success in related fields. Thus, in this paper we propose a NF framework to directly model the enhancement process by density estimation of clean speech utterances conditioned on their noisy counterpart. The WaveGlow model from speech synthesis is adapted to enable direct enhancement of noisy utterances in time domain. In addition, we demonstrate that nonlinear input companding benefits the model performance by equalizing the distribution of input samples. Experimental evaluation on a publicly available dataset shows comparable results to current state-of-the-art GAN-based approaches, while surpassing the chosen baselines using objective evaluation metrics.

arXiv commentAccepted to ICASSP 2021

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Normalizing Flows: An Introduction and Review of Current Methods
• Weight Normalization: A Simple Reparameterization to Accelerate Training of Deep Neural Networks
• Density estimation using Real NVP
• Normalizing Flows for Probabilistic Modeling and Inference
• Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design
• Improving GANs for Speech Enhancement
• SqueezeWave: Extremely Lightweight Vocoders for On-device Speech Synthesis
• Speech Enhancement using Self-Adaptation and Multi-Head Self-Attention

Cited by in corpus (3)

• A Study on Speech Enhancement Based on Diffusion Probabilistic Model
• SEFGAN: Harvesting the Power of Normalizing Flows and GANs for Efficient High-Quality Speech Enhancement
• INN-PAR: Invertible Neural Network for PPG to ABP Reconstruction