Author:
Cabañas-Molero Pablo,Muñoz-Montoro Antonio J.,Vera-Candeas Pedro,Ranilla José
Abstract
AbstractClassical music, unlike popular music, is usually recorded live with close microphone techniques. For this reason, isolated tracks are not available to create the final mixture/stream, and so the mixing process requires greater effort. Source separation methods are a potential solution to this problem. However, current algorithms are not fast enough to yield real-time separation in professional setups with dozens of microphones and sources. In this paper, we propose a fast approach consisting of a panning-based multichannel non-negative matrix factorization model to separate classical music. We tested the system on real professional recordings, where we were able to reach real-time with very low latency and promising quality.
Funder
Regional Government of Andalucia
“Ministerio de Ciencia e Innovacion” of Spain
Universidad de Jaén
Publisher
Springer Science and Business Media LLC
Subject
Hardware and Architecture,Information Systems,Theoretical Computer Science,Software
Reference20 articles.
1. Stöter FR, Uhlich S, Liutkus A, Mitsufuji Y (2019) Open-unmix - a reference implementation for music source separation. J Open Sour Softw 4(41):1667. https://doi.org/10.21105/joss.01667
2. Hennequin R, Khlif A, Voituret F, Moussallam M (2020) Spleeter: a fast and efficient music source separation tool with pre-trained models. J Open Sour Softw 5(50):2154
3. Défossez A, Usunier N, Bottou L, Bach F Demucs: deep extractor for music sources with extra unlabeled data remixed. Available from: arXiv:1909.01174
4. Huber DM, Runstein R (2013) Modern recording techniques. Routledge;
5. Kokkinis EK, Mourjopoulos J (2010) Unmixing acoustic sources in real reverberant environments for close-microphone applications. J Audio Eng Soc 58(11):907–922