Class-conditional Embeddings for Music Source Separation

Class-conditional Embeddings for Music Source Separation
复制标题

用于音乐源分离的类条件嵌入

DOI:
10.1109/icassp.2019.8683007
复制
发表时间:
2018
期刊:
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
影响因子:
--
通讯作者:
Jonathan Le Roux
Jonathan Le Roux
中科院分区:
--
文献类型:
--
作者:
Prem Seetharaman;G. Wichern;Shrikant Venkataramani;Jonathan Le Roux

文献摘要

参考文献

被引文献

相似文献

在音乐混合中分离单个乐器具有无数潜在的应用,并且考虑到最近深度学习方法所达到的性能水平,似乎可以立即实现。虽然大多数音乐源分离技术为每个乐器学习一个独立的模型,但我们建议在受深度聚类和深度吸引子网络启发的混合中,为所有乐器的时频箱使用一个共同的嵌入空间。此外,使用辅助网络生成高斯混合模型(GMM)的参数,其中嵌入空间中GMM分量的后验分布可用于创建将单个源从混合物中分离出来的掩模。除了在MUSDB-18数据集上优于掩码推理基线之外,我们的嵌入空间很容易解释,可以用于基于查询的分离。
Isolating individual instruments in a musical mixture has a myriad of potential applications, and seems imminently achievable given the levels of performance reached by recent deep learning methods. While most musical source separation techniques learn an independent model for each instrument, we propose using a common embedding space for the time-frequency bins of all instruments in a mixture inspired by deep clustering and deep attractor networks. Additionally, an auxiliary network is used to generate parameters of a Gaussian mixture model (GMM) where the posterior distribution over GMM components in the embedding space can be used to create a mask that separates individual sources from a mixture. In addition to outperforming a mask-inference baseline on the MUSDB-18 dataset, our embedding space is easily interpretable and can be used for query-based separation.
DOI: 10.1109/taslp.2018.2858559
发表时间: 2018-04
期刊: IEEE/ACM Transactions on Audio, Speech, and Language Processing
影响因子: --
作者:
Brian McFee;J. Salamon;J. Bello
通讯作者: Brian McFee;J. Salamon;J. Bello
DOI: 10.1109/waspaa.2017.8170052
发表时间: 2017-10
期刊: 2017 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA)
影响因子: --
作者:
J. Salamon;D. MacConnell;M. Cartwright;P. Li;J. Bello
通讯作者: J. Salamon;D. MacConnell;M. Cartwright;P. Li;J. Bello