Saved in:
Bibliographic Details
Main Authors: Zhao, Haixin, Yang, Kaixuan, Madhu, Nilesh
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.11395
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • To further reduce the complexity of lightweight speech enhancement models, we introduce a gating-based Dynamically Slimmable Network (DSN). The DSN comprises static and dynamic components. For architecture-independent applicability, we introduce distinct dynamic structures targeting the commonly used components, namely, grouped recurrent neural network units, multi-head attention, convolutional, and fully connected layers. A policy module adaptively governs the use of dynamic parts at a frame-wise resolution according to the input signal quality, controlling computational load. We further propose Metric-Guided Training (MGT) to explicitly guide the policy module in assessing input speech quality. Experimental results demonstrate that the DSN achieves comparable enhancement performance in instrumental metrics to the state-of-the-art lightweight baseline, while using only 73% of its computational load on average. Evaluations of dynamic component usage ratios indicate that the MGT-DSN can appropriately allocate network resources according to the severity of input signal distortion.