# RaD-Net 2: A causal two-stage repairing and denoising speech enhancement network with knowledge distillation and complex axial self-attention

arXiv: https://arxiv.org/abs/2406.07498 · HTML: https://mlsanity.com/paper/2406.07498/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Speech and audio distillation, one of 12 directions in Knowledge distillation.
- Builds on 17 of its 26 references that are in the map.
- First posted on arXiv 11 Jun 2024.

## Placed in

- Knowledge distillation / [Speech and audio distillation](https://mlsanity.com/distill/speech-and-audio-distillation/)

## Builds on (most cited first, 17 of 17)

- [HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis](https://mlsanity.com/paper/2010.05646/) (2020)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation](https://mlsanity.com/paper/1809.07454/) (2018)
- [DNSMOS P.835: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressors](https://mlsanity.com/paper/2110.01763/) (2021)
- [DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement](https://mlsanity.com/paper/2008.00264/) (2020)
- [UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation](https://mlsanity.com/paper/2106.07889/) (2021)
- [FRCRN: Boosting Feature Representation using Frequency Recurrence for Monaural Speech Enhancement](https://mlsanity.com/paper/2206.07293/) (2022)
- [Fast Spectrogram Inversion using Multi-head Convolutional Neural Networks](https://mlsanity.com/paper/1808.06719/) (2018)
- [ICASSP 2024 Speech Signal Improvement Challenge](https://mlsanity.com/paper/2401.14444/) (2024)
- [S-DCCRN: Super Wide Band DCCRN with learnable complex feature for speech enhancement](https://mlsanity.com/paper/2111.08387/) (2021)
- [Supervised and Unsupervised Speech Enhancement Using Nonnegative Matrix Factorization](https://mlsanity.com/paper/1709.05362/) (2017)
- [ICASSP 2023 Speech Signal Improvement Challenge](https://mlsanity.com/paper/2303.06566/) (2023)
- [KS-Net: Multi-band joint speech restoration and enhancement network for 2024 ICASSP SSI Challenge](https://mlsanity.com/paper/2402.01808/) (2024)
- [Gesper: A Restoration-Enhancement Framework for General Speech Reconstruction](https://mlsanity.com/paper/2306.08454/) (2023)
- [RaD-Net: A Repairing and Denoising Network for Speech Signal Improvement](https://mlsanity.com/paper/2401.04389/) (2024)
- [Two-stage Neural Network for ICASSP 2023 Speech Signal Improvement Challenge](https://mlsanity.com/paper/2303.07621/) (2023)
- [D2Former: A Fully Complex Dual-Path Dual-Decoder Conformer Network using Joint Complex Masking and Complex Spectral Mapping for Monaural Speech Enhancement](https://mlsanity.com/paper/2302.11832/) (2023)

## Built on by (most cited first, 1 of 1)

- [Knowledge Distillation for Speech Denoising by Latent Representation Alignment with Cosine Distance](https://mlsanity.com/paper/2505.03442/) (2025)

## Abstract (arXiv)

In real-time speech communication systems, speech signals are often degraded by multiple distortions. Recently, a two-stage Repair-and-Denoising network (RaD-Net) was proposed with superior speech quality improvement in the ICASSP 2024 Speech Signal Improvement (SSI) Challenge. However, failure to use future information and constraint receptive field of convolution layers limit the system's performance. To mitigate these problems, we extend RaD-Net to its upgraded version, RaD-Net 2. Specifically, a causality-based knowledge distillation is introduced in the first stage to use future information in a causal way. We use the non-causal repairing network as the teacher to improve the performance of the causal repairing network. In addition, in the second stage, complex axial self-attention is applied in the denoising network's complex feature encoder/decoder. Experimental results on the ICASSP 2024 SSI Challenge blind test set show that RaD-Net 2 brings 0.10 OVRL DNSMOS improvement compared to RaD-Net.