# I-DCCRN-VAE: An Improved Deep Representation Learning Framework for Complex VAE-based Single-channel Speech Enhancement

arXiv: https://arxiv.org/abs/2510.12485 · HTML: https://mlsanity.com/paper/2510.12485/

## Key facts

- Cited by 1 papers in the map (data through 19 Oct 2026).
- Placed in Speech enhancement and noise suppression, one of 11 directions in Speech and audio language models.
- Builds on 17 of its 25 references that are in the map.
- First posted on arXiv 14 Oct 2025.

## Placed in

- Speech and audio language models / [Speech enhancement and noise suppression](https://mlsanity.com/speech/speech-enhancement-and-noise-suppression/)

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

- [DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement](https://mlsanity.com/paper/2008.00264/) (2020)
- [SEGAN: Speech Enhancement Generative Adversarial Network](https://mlsanity.com/paper/1703.09452/) (2017)
- [Speech Enhancement and Dereverberation with Diffusion-based Generative Models](https://mlsanity.com/paper/2208.05830/) (2022)
- [Conditional Diffusion Probabilistic Model for Speech Enhancement](https://mlsanity.com/paper/2202.05256/) (2022)
- [Interspeech 2021 Deep Noise Suppression Challenge](https://mlsanity.com/paper/2101.01902/) (2021)
- [CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancement](https://mlsanity.com/paper/2209.11112/) (2022)
- [Variational Autoencoder for Speech Enhancement with a Noise-Aware Encoder](https://mlsanity.com/paper/2102.08706/) (2021)
- [Unsupervised Speech Enhancement using Dynamical Variational Auto-Encoders](https://mlsanity.com/paper/2106.12271/) (2021)
- [Improving GANs for Speech Enhancement](https://mlsanity.com/paper/2001.05532/) (2020)
- [Unsupervised speech enhancement with diffusion-based generative models](https://mlsanity.com/paper/2309.10450/) (2023)
- [Investigating the Design Space of Diffusion Models for Speech Enhancement](https://mlsanity.com/paper/2312.04370/) (2023)
- [A Deep Representation Learning-based Speech Enhancement Method Using Complex Convolution Recurrent Variational Autoencoder](https://mlsanity.com/paper/2312.09620/) (2023)
- [A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder](https://mlsanity.com/paper/2201.09875/) (2022)
- [A deep representation learning speech enhancement method using β-VAE](https://mlsanity.com/paper/2205.05581/) (2022)
- [Fast and efficient speech enhancement with variational autoencoders](https://mlsanity.com/paper/2211.02728/) (2022)
- [A Two-Stage Deep Representation Learning-Based Speech Enhancement Method Using Variational Autoencoder and Adversarial Training](https://mlsanity.com/paper/2211.09166/) (2022)
- [Investigation of Speech and Noise Latent Representations in Single-channel VAE-based Speech Enhancement](https://mlsanity.com/paper/2508.05293/) (2025)

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

- [U-PAST: A Phase-Aware Audio Spectrogram Transformer-U-Net for Single-Channel Speech Enhancement](https://mlsanity.com/paper/2609.00431/) (2026)

## Abstract (arXiv)

Recently, a complex variational autoencoder (VAE)-based single-channel speech enhancement system based on the DCCRN architecture has been proposed. In this system, a noise suppression VAE (NSVAE) learns to extract clean speech representations from noisy speech using pretrained clean speech and noise VAEs with skip connections. In this paper, we improve DCCRN-VAE by incorporating three key modifications: 1) removing the skip connections in the pretrained VAEs to encourage more informative speech and noise latent representations; 2) using β-VAE in pretraining to better balance reconstruction and latent space regularization; and 3) a NSVAE generating both speech and noise latent representations. Experiments show that the proposed system achieves comparable performance as the DCCRN and DCCRN-VAE baselines on the matched DNS3 dataset but outperforms the baselines on mismatched datasets (WSJ0-QUT, Voicebank-DEMEND), demonstrating improved generalization ability. In addition, an ablation study shows that a similar performance can be achieved with classical fine-tuning instead of adversarial training, resulting in a simpler training pipeline.