UNMIXX: Untangling Highly Correlated Singing Voices Mixtures
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917209735954432 |
|---|---|
| author | Jung, Jihoo Kim, Ji-Hoon Kwak, Doyeop Lee, Junwon Nam, Juhan Chung, Joon Son |
| author_facet | Jung, Jihoo Kim, Ji-Hoon Kwak, Doyeop Lee, Junwon Nam, Juhan Chung, Joon Son |
| contents | We introduce UNMIXX, a novel framework for multiple singing voices separation (MSVS). While related to speech separation, MSVS faces unique challenges: data scarcity and the highly correlated nature of singing voices mixture. To address these issues, we propose UNMIXX with three key components: (1) musically informed mixing strategy to construct highly correlated, music-like mixtures, (2) cross-source attention that drives representations of two singers apart via reverse attention, and (3) magnitude penalty loss penalizing erroneously assigned interfering energy. UNMIXX not only addresses data scarcity by simulating realistic training data, but also excels at separating highly correlated mixtures through cross-source interactions at both the architectural and loss levels. Our extensive experiments demonstrate that UNMIXX greatly enhances performance, with SDRi gains exceeding 2.2 dB over prior work. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12802 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UNMIXX: Untangling Highly Correlated Singing Voices Mixtures Jung, Jihoo Kim, Ji-Hoon Kwak, Doyeop Lee, Junwon Nam, Juhan Chung, Joon Son Sound Audio and Speech Processing We introduce UNMIXX, a novel framework for multiple singing voices separation (MSVS). While related to speech separation, MSVS faces unique challenges: data scarcity and the highly correlated nature of singing voices mixture. To address these issues, we propose UNMIXX with three key components: (1) musically informed mixing strategy to construct highly correlated, music-like mixtures, (2) cross-source attention that drives representations of two singers apart via reverse attention, and (3) magnitude penalty loss penalizing erroneously assigned interfering energy. UNMIXX not only addresses data scarcity by simulating realistic training data, but also excels at separating highly correlated mixtures through cross-source interactions at both the architectural and loss levels. Our extensive experiments demonstrate that UNMIXX greatly enhances performance, with SDRi gains exceeding 2.2 dB over prior work. |
| title | UNMIXX: Untangling Highly Correlated Singing Voices Mixtures |
| topic | Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2601.12802 |