GAN-Based Speech Enhancement for Low SNR Using Latent Feature Conditioning

Shetu SS, Habets EA, Brendel A (2025)


Publication Type: Conference contribution

Publication year: 2025

Journal

Publisher: Institute of Electrical and Electronics Engineers Inc.

Conference Proceedings Title: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

Event location: Hyderabad, IND

DOI: 10.1109/ICASSP49660.2025.10890549

Abstract

Enhancing speech quality under adverse SNR conditions remains a significant challenge for discriminative deep neural network (DNN)-based approaches. In this work, we propose DisCoGAN, which is a time-frequency-domain generative adversarial network (GAN) conditioned by the latent features of a discriminative model pre-trained for speech enhancement in low SNR scenarios. Our proposed method achieves superior performance compared to state-of-the-art discriminative methods and also surpasses end-to-end (E2E) trained GAN models. We also investigate the impact of various configurations for conditioning the proposed GAN model with the discriminative model and assess their influence on enhancing speech quality.

How to cite

APA:

Shetu, S.S., Habets, E.A., & Brendel, A. (2025). GAN-Based Speech Enhancement for Low SNR Using Latent Feature Conditioning. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Hyderabad, IND: Institute of Electrical and Electronics Engineers Inc..

MLA:

Shetu, Shrishti Saha, Emanuël A.P. Habets, and Andreas Brendel. "GAN-Based Speech Enhancement for Low SNR Using Latent Feature Conditioning." Proceedings of the 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025, Hyderabad, IND Institute of Electrical and Electronics Engineers Inc., 2025.

BibTeX: Download