Boosting Objective Scores of a Speech Enhancement Model by MetricGAN Post-processing

Abstract

The Transformer architecture has demonstrated a superior ability compared to recurrent neural networks in many different natural language processing applications. Therefore, our study applies a modified Transformer in a speech enhancement task. Specifically, positional encoding in the Transformer may not be necessary for speech enhancement, and hence, it is replaced by convolutional layers. To further improve the perceptual evaluation of the speech quality (PESQ) scores of enhanced speech, the L_1 pre-trained Transformer is fine-tuned using a MetricGAN framework. The proposed MetricGAN can be treated as a general post-processing module to further boost the objective scores of interest. The experiments were conducted using the data sets provided by the organizer of the Deep Noise Suppression (DNS) challenge. Experimental results demonstrated that the proposed system outperformed the challenge baseline, in both subjective and objective evaluations, with a large margin.

arXiv commentAccepted by APSIPA 2020

References in corpus (5)

• Spectral Normalization for Generative Adversarial Networks
• Convolutional Sequence to Sequence Learning
• MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement
• Learning with Learned Loss Function: Speech Enhancement with Quality-Net to Improve Perceptual Evaluation of Speech Quality
• The INTERSPEECH 2020 Deep Noise Suppression Challenge: Datasets, Subjective Speech Quality and Testing Framework

Cited by in corpus (2)

• Unsupervised Noise Adaptive Speech Enhancement by Discriminator-Constrained Optimal Transport
• CITISEN: A Deep Learning-Based Speech Signal-Processing Mobile Application