# HD-DEMUCS: General Speech Restoration with Heterogeneous Decoders

arXiv: https://arxiv.org/abs/2306.01411 · HTML: https://mlsanity.com/paper/2306.01411/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jun 2023.

## Built on by (most cited first, 2 of 2)

- [EDNet: A Versatile Speech Enhancement Framework with Gating Mamba Mechanism and Phase Shift-Invariant Training](https://mlsanity.com/paper/2506.16231/) (2025)
- [A Dual-Branch Parallel Network for Speech Enhancement and Restoration](https://mlsanity.com/paper/2409.08702/) (2024)

## Abstract (arXiv)

This paper introduces an end-to-end neural speech restoration model, HD-DEMUCS, demonstrating efficacy across multiple distortion environments. Unlike conventional approaches that employ cascading frameworks to remove undesirable noise first and then restore missing signal components, our model performs these tasks in parallel using two heterogeneous decoder networks. Based on the U-Net style encoder-decoder framework, we attach an additional decoder so that each decoder network performs noise suppression or restoration separately. We carefully design each decoder architecture to operate appropriately depending on its objectives. Additionally, we improve performance by leveraging a learnable weighting factor, aggregating the two decoder output waveforms. Experimental results with objective metrics across various environments clearly demonstrate the effectiveness of our approach over a single decoder or multi-stage systems for general speech restoration task.