MAGE: A Coarse-to-Fine Speech Enhancer with Masked Generative Model
Abstract
Speech enhancement remains challenging due to the trade-off between efficiency and perceptual quality. In this paper, we introduce MAGE, a Masked Audio Generative Enhancer that advances generative speech enhancement through a compact and robust design. Unlike prior masked generative models with random masking, MAGE employs a scarcity-aware coarse-to-fine masking strategy that prioritizes frequent tokens in early steps and rare tokens in later refinements, improving efficiency and generalization. We also propose a lightweight corrector module that further stabilizes inference by detecting low-confidence predictions and re-masking them for refinement. Built on BigCodec and finetuned from Qwen2.5-0.5B, MAGE is reduced to 200M parameters through selective layer retention. Experiments on DNS Challenge and noisy LibriSpeech show that MAGE achieves state-of-the-art perceptual quality and significantly reduces word error rate for downstream recognition, outperforming larger baselines. Audio examples are available at https://hieugiaosu.github.io/MAGE/.
Cite
@article{arxiv.2509.19881,
title = {MAGE: A Coarse-to-Fine Speech Enhancer with Masked Generative Model},
author = {The Hieu Pham and Tan Dat Nguyen and Phuong Thanh Tran and Joon Son Chung and Duc Dung Nguyen},
journal= {arXiv preprint arXiv:2509.19881},
year = {2026}
}
Comments
ICASSP 2026