Raw Waveform-based Speech Enhancement by Fully Convolutional Networks

Abstract

This study proposes a fully convolutional network (FCN) model for raw waveform-based speech enhancement. The proposed system performs speech enhancement in an end-to-end (i.e., waveform-in and waveform-out) manner, which dif-fers from most existing denoising methods that process the magnitude spectrum (e.g., log power spectrum (LPS)) only. Because the fully connected layers, which are involved in deep neural networks (DNN) and convolutional neural networks (CNN), may not accurately characterize the local information of speech signals, particularly with high frequency components, we employed fully convolutional layers to model the waveform. More specifically, FCN consists of only convolutional layers and thus the local temporal structures of speech signals can be efficiently and effectively preserved with relatively few weights. Experimental results show that DNN- and CNN-based models have limited capability to restore high frequency components of waveforms, thus leading to decreased intelligibility of enhanced speech. By contrast, the proposed FCN model can not only effectively recover the waveforms but also outperform the LPS-based DNN baseline in terms of short-time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ). In addition, the number of model parameters in FCN is approximately only 0.2% compared with that in both DNN and CNN.

References in corpus (2)

• WaveNet: A Generative Model for Raw Audio
• Very Deep Convolutional Neural Networks for Raw Waveforms

Cited by in corpus (10)

• Supervised Speech Separation Based on Deep Learning: An Overview
• Time-Domain Multi-modal Bone/air Conducted Speech Enhancement
• Generative adversarial network-based glottal waveform model for statistical parametric speech synthesis
• End-to-End Speech Separation with Unfolded Iterative Phase Reconstruction
• End-to-End Waveform Utterance Enhancement for Direct Evaluation Metrics Optimization by Fully Convolutional Neural Networks
• End-to-end Networks for Supervised Single-channel Speech Separation
• A Fully Convolutional Neural Network Approach to End-to-End Speech Enhancement
• Convolutional Neural Networks to Enhance Coded Speech
• A study on speech enhancement using exponent-only floating point quantized neural network (EOFP-QNN)
• Learning Metrics from Mean Teacher: A Supervised Learning Method for Improving the Generalization of Speaker Verification System