Static and Dynamic Source Separation Using Nonnegative Factorizations: A unified view

167 citations recorded by the source corpus.

How researchers see this paper

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

Speech Denoising with Deep Feature Losses (2018, methodology): “Earlier denoising systems relied on spectrogram-domain signal processing methods [1], followed more recently by methods based on spectrogram factorization [3].”
Music Source Separation in the Waveform Domain (2019, methodology): “Non-negative matrix factorization techniques [Smaragdis et al., 2014] model the power spectrum as a weighted sum of a learnt spectral dictionary, whose elements are grouped into individual sources.”
Singing Voice Separation with Deep U-Net Convolutional Networks (2017, methodology): “Successful results have been achieved with non-negative matrix factorization [26, 30, 32], Bayesian methods [21], and the analysis of repeating structures [23].”
Equalization matching of speech recordings in real-world environments (2016, methodology): “Our best guess is to compute the filter G as the ratio: G(ω) = Ya(ω)/Yb(ω) (8)”
Generative adversarial network-based approach to signal reconstruction from magnitude spectrogram (2018, methodology): “In fact, many methods for monaural audio source separation are applied to magnitude spectrograms [1]–[3].”