{"ID":2880188,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.14556","arxiv_id":"2508.14556","title":"Mamba2 Meets Silence: Robust Vocal Source Separation for Sparse Regions","abstract":"We introduce a new music source separation model tailored for accurate vocal isolation. Unlike Transformer-based approaches, which often fail to capture intermittently occurring vocals, our model leverages Mamba2, a recent state space model, to better capture long-range temporal dependencies. To handle long input sequences efficiently, we combine a band-splitting strategy with a dual-path architecture. Experiments show that our approach outperforms recent state-of-the-art models, achieving a cSDR of 11.03 dB-the best reported to date-and delivering substantial gains in uSDR. Moreover, the model exhibits stable and consistent performance across varying input lengths and vocal occurrence patterns. These results demonstrate the effectiveness of Mamba-based models for high-resolution audio processing and open up new directions for broader applications in audio research.","short_abstract":"We introduce a new music source separation model tailored for accurate vocal isolation. Unlike Transformer-based approaches, which often fail to capture intermittently occurring vocals, our model leverages Mamba2, a recent state space model, to better capture long-range temporal dependencies. To handle long input seque...","url_abs":"https://arxiv.org/abs/2508.14556","url_pdf":"https://arxiv.org/pdf/2508.14556v2","authors":"[\"Euiyeon Kim\",\"Yong-Hoon Choi\"]","published":"2025-08-20T09:19:11Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\",\"eess.AS\"]","methods":"[\"Transformer\"]","has_code":false}