pith. sign in

arxiv: 2110.09924 · v2 · pith:HZM4QNIFnew · submitted 2021-10-19 · 📡 eess.AS · cs.SD

Speech Enhancement Based on Cyclegan with Noise-informed Training

classification 📡 eess.AS cs.SD
keywords cyclegantrainingapproachclean-to-noisysystemconversionenhancementnoise-informed
0
0 comments X
read the original abstract

Cycle-consistent generative adversarial networks (CycleGAN) were successfully applied to speech enhancement (SE) tasks with unpaired noisy-clean training data. The CycleGAN SE system adopted two generators and two discriminators trained with losses from noisy-to-clean and clean-to-noisy conversions. CycleGAN showed promising results for numerous SE tasks. Herein, we investigate a potential limitation of the clean-to-noisy conversion part and propose a novel noise-informed training (NIT) approach to improve the performance of the original CycleGAN SE system. The main idea of the NIT approach is to incorporate target domain information for clean-to-noisy conversion to facilitate a better training procedure. The experimental results confirmed that the proposed NIT approach improved the generalization capability of the original CycleGAN SE system with a notable margin.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.