Saved in:
Bibliographic Details
Main Authors: Tan, Shihong, Wang, Haoyu, Ni, Youran, Hou, Yingzhao, Luo, Jiayue, Hu, Zipei, Dou, Han, Han, Zerui, Pan, Ningning, Wang, Yuzhu, Huang, Gongping
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.19825
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908859964063744
author Tan, Shihong
Wang, Haoyu
Ni, Youran
Hou, Yingzhao
Luo, Jiayue
Hu, Zipei
Dou, Han
Han, Zerui
Pan, Ningning
Wang, Yuzhu
Huang, Gongping
author_facet Tan, Shihong
Wang, Haoyu
Ni, Youran
Hou, Yingzhao
Luo, Jiayue
Hu, Zipei
Dou, Han
Han, Zerui
Pan, Ningning
Wang, Yuzhu
Huang, Gongping
contents Music source restoration (MSR) aims to recover unprocessed stems from mixed and mastered recordings. The challenge lies in both separating overlapping sources and reconstructing signals degraded by production effects such as compression and reverberation. We therefore propose DTT-BSR, a hybrid generative adversarial network (GAN) combining rotary positional embeddings (RoPE) transformer for long-term temporal modeling with dual-path band-split recurrent neural network (RNN) for multi-resolution spectral processing. Our model achieved 3rd place on the objective leaderboard and 4th place on the subjective leaderboard on the ICASSP 2026 MSR Challenge, demonstrating exceptional generation fidelity and semantic alignment with a compact size of 7.1M parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2602_19825
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle DTT-BSR: GAN-based DTTNet with RoPE Transformer Enhancement for Music Source Restoration
Tan, Shihong
Wang, Haoyu
Ni, Youran
Hou, Yingzhao
Luo, Jiayue
Hu, Zipei
Dou, Han
Han, Zerui
Pan, Ningning
Wang, Yuzhu
Huang, Gongping
Audio and Speech Processing
Sound
Music source restoration (MSR) aims to recover unprocessed stems from mixed and mastered recordings. The challenge lies in both separating overlapping sources and reconstructing signals degraded by production effects such as compression and reverberation. We therefore propose DTT-BSR, a hybrid generative adversarial network (GAN) combining rotary positional embeddings (RoPE) transformer for long-term temporal modeling with dual-path band-split recurrent neural network (RNN) for multi-resolution spectral processing. Our model achieved 3rd place on the objective leaderboard and 4th place on the subjective leaderboard on the ICASSP 2026 MSR Challenge, demonstrating exceptional generation fidelity and semantic alignment with a compact size of 7.1M parameters.
title DTT-BSR: GAN-based DTTNet with RoPE Transformer Enhancement for Music Source Restoration
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2602.19825