[2002.07016] Meta-learning Extractors for Music Source Separationopen searchopen navigation menucontact arXivarXiv Twitter

We propose a hierarchical meta-learning-inspired model for music source separation (Meta-TasNet) in which a generator model is used to predict the weights of individual extractor models. This enables efficient parameter-sharing, while still allowing for instrument-specific parameterization. Meta-TasNet is shown to be more effective than the models trained independently or in a multi-task setting, and achieve performance comparable with state-of-the-art methods. In comparison to the latter, our extractors contain fewer parameters and have faster run-time performance. We discuss important architectural considerations, and explore the costs and benefits of this approach.

1 mentions: @keunwoochoi
Date: 2020/05/18 02:21

Related Entries

Read more GitHub - keunwoochoi/openmic-2018-tfrecord
0 users, 1 mentions 2020/04/16 21:51
Read more [1910.13069] Disentangling Timbre and Singing Style with Multi-singer Singing Synthesis Systemopen s...
0 users, 1 mentions 2020/05/18 02:21
Read more NNlib · Flux
0 users, 1 mentions 2020/08/14 09:51
Read more Kapre — Kapre 2017 documentation
0 users, 1 mentions 2020/08/27 21:52
Read more GitHub - keunwoochoi/kapre at kapre-0.3.2
0 users, 1 mentions 2020/08/30 14:21