Multichannel music separation with deep neural networks

By Aditya Arie Nugraha, Antoine Liutkus, Emmanuel Vincent. 2016

87 citations recorded by the source corpus.

Subfields: Computational Mechanics, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Signal Processing

Citations in the corpus

Music Source Separation in the Waveform Domain (2019, methodology): “MMDenseLSTM, D3Net, Spleeter, and Open Unmix use Wiener filtering [Nugraha et al., 2016] as a last post processing step.”
Denoising Auto-Encoder with Recurrent Skip Connections and Residual Regression for Music Source Separation (2018, methodology): “The models are implemented with PyTorch.2 We will report the evaluation result in terms of signal-to-distortion ratio (SDR) [14], as it is the most widely used metric in literature [13], [15], [16] 1https://github.com/sigsep/sigsep-mus-eval and https://github.com/sigsep/ sigsep-mus-2018-analysis 2ht”
Mmdenselstm: An Efficient Combination of Convolutional and Recurrent Neural Networks for Audio Source Separation (2018, methodology): “son of MMDenseLSTM conﬁgurations. type Sa Sb P SDR 2.83 2.31 2.47 Table 3. Comparison of SDR on DSD100 dataset. SDR in dB Method Bass Drums Other Vocals Acco. DeepNMF [4] 1.88 2.11 2.64 2.75 8.90 NUG [10] 2.72 3.89 3.18 4.55 10.29 BLSTM [12] 2.89 4.00 3.24 4.86 11.26 BLEND [12] 2.98 4.13 3.52 5.23 1”
Multi-Scale multi-band densenets for audio source separation (2017, methodology): “This approach was referred as NUG1 in [10].”
An analysis of environment, microphone and data simulation mismatches in robust speech recognition (2017, unknown): “We considered the DNN-based multichannel source separation technique of Nugraha et al. (2016a).”
D3Net: Densely connected multidilated DenseNet for music source separation (2020, background): “Many types of neural network architecture have been proposed, including feedforward fully connected networks (FNNs) [1, 2], recurrent neural networks (RNNs) [3], convolutional neural networks (CNNs) [4–9] and their combinations [10, 11].”
Singing Voice Separation: A Study on Training Data (2019, methodology): “In the previous literature [10, 11, 12, 13, 9] different architectures are usually compared using the same train/test datasets, but to the best of our knowledge, there are no previous works that study particularly the inﬂuence of these datasets.”