# Mel-Band RoFormer for Music Source Separation

arXiv: https://arxiv.org/abs/2310.01809 · HTML: https://mlsanity.com/paper/2310.01809/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Oct 2023.

## Built on by (most cited first, 17 of 17)

- [Source Separation by Flow Matching](https://mlsanity.com/paper/2505.16119/) (2025)
- [SingMOS-Pro: An Comprehensive Benchmark for Singing Quality Assessment](https://mlsanity.com/paper/2510.01812/) (2025)
- [Music Source Restoration](https://mlsanity.com/paper/2505.21827/) (2025)
- [SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis](https://mlsanity.com/paper/2602.07803/) (2026)
- [An Ensemble Approach to Music Source Separation: A Comparative Analysis of Conventional and Hierarchical Stem Separation](https://mlsanity.com/paper/2410.20773/) (2024)
- [Chirp Localization via Fine-Tuned Transformer Model: A Proof-of-Concept Study](https://mlsanity.com/paper/2503.22713/) (2025)
- [R2-SVC: Towards Real-World Robust and Expressive Zero-shot Singing Voice Conversion](https://mlsanity.com/paper/2510.20677/) (2025)
- [Summary of The Inaugural Music Source Restoration Challenge](https://mlsanity.com/paper/2601.04343/) (2026)
- [Modeling Music as a Time-Frequency Image: A 2D Tokenizer for Music Generation](https://mlsanity.com/paper/2605.15831/) (2026)
- [TISDiSS: A Training-Time and Inference-Time Scalable Framework for Discriminative Source Separation](https://mlsanity.com/paper/2509.15666/) (2025)
- [Source Separation for A Cappella Music](https://mlsanity.com/paper/2509.26580/) (2025)
- [Persian MusicGen: A Large-Scale Dataset and Culturally-Aware Generative Model for Persian Music](https://mlsanity.com/paper/2605.14765/) (2026)
- [SURF: Separation via Unsupervised Remixing Flow](https://mlsanity.com/paper/2606.04921/) (2026)
- [HoliDubber: Holistic Video Dubbing for Complex Acoustic Scenes via Text-Guided Audio Synthesis](https://mlsanity.com/paper/2606.09098/) (2026)
- [Music-Source-Separation-Training (MSST): A Unified Framework for Training and Evaluating Music Demixing Models](https://mlsanity.com/paper/2607.23395/) (2026)
- [VocalRender: Score-Native Singing Voice Synthesis for Real-World Composition](https://mlsanity.com/paper/2607.27768/) (2026)
- [Iterative Audio Separation with Mixture Consistency via MIMO Model Extension](https://mlsanity.com/paper/2609.07226/) (2026)

## Abstract (arXiv)

Recently, multi-band spectrogram-based approaches such as Band-Split RNN (BSRNN) have demonstrated promising results for music source separation. In our recent work, we introduce the BS-RoFormer model which inherits the idea of band-split scheme in BSRNN at the front-end, and then uses the hierarchical Transformer with Rotary Position Embedding (RoPE) to model the inner-band and inter-band sequences for multi-band mask estimation. This model has achieved state-of-the-art performance, but the band-split scheme is defined empirically, without analytic supports from the literature. In this paper, we propose Mel-RoFormer, which adopts the Mel-band scheme that maps the frequency bins into overlapped subbands according to the mel scale. In contract, the band-split mapping in BSRNN and BS-RoFormer is non-overlapping and designed based on heuristics. Using the MUSDB18HQ dataset for experiments, we demonstrate that Mel-RoFormer outperforms BS-RoFormer in the separation tasks of vocals, drums, and other stems.