DTT-BSR: GAN-BASED DTTNET WITH ROPE TRANSFORMER ENHANCEMENT FOR MUSIC SOURCE RESTORATION
Shihong Tan, Haoyu Wang, Youran Ni, Zerui Han, Ningning Pan, Yuzhu Wang, Gongping Huang
Abstract
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.
BibTeX
@inproceedings{icassp2026_dttbsrganbaseddt,
title = {DTT-BSR: GAN-BASED DTTNET WITH ROPE TRANSFORMER ENHANCEMENT FOR MUSIC SOURCE RESTORATION},
author = {Shihong Tan and Haoyu Wang and Youran Ni and Zerui Han and Ningning Pan and Yuzhu Wang and Gongping Huang},
booktitle = {ICASSP 2026},
year = {2026}
}