Single channel speech music separation using nonnegative matrix factorization and spectral masks

By Emad M. Grais, Hakan Erdogan. 2011

97 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

1. Sabine Süsstrunk
2. Edo Collins
3. Radhakrishna Achanta
4. M. U. Sen
5. Hakan Erdogan
6. Emad M. Grais
7. Siavash Arjomand Bigdeli

In Signal Processing

1. Sabine Süsstrunk
2. Edo Collins
3. Siavash Arjomand Bigdeli
4. M. U. Sen
5. Radhakrishna Achanta
6. Hakan Erdogan
7. Emad M. Grais
8. Deliang Wang

Citations in the corpus

• SEANet: A Multi-modal Speech Enhancement Network (2020, neutral): “Traditional signal processing methods adopt spectral noise subtraction [1, 2], spectral masking [3, 4], statistical methods based on Wiener filtering [5] and Bayesian estimators [6, 7].”
• Deep neural networks for single channel source separation (2013, methodology): “Another novelty of this paper is using nonnegative matrix factorization [23] to find initial estimates for the sources rather than using random initialization.”
• Deep Feature Factorization For Concept Discovery (2018, methodology): “use labeled data for evaluation only. 2.3 Non-negative Matrix Factorization Non-negative matrix factorization (NMF) has been used to analyze data from various domains, such as audio source separation [9], document clustering [10] and face recognition [11]. There has been work extending NMF to multip”
• Detecting Memorization in ReLU Networks (2018, background): “…NMF has gained popularity due to its producing meaningful factorizations that lend themselves to qualitative interpretation across various domains such as document clustering (Xu et al., 2003), audio source separation (Grais & Erdogan, 2011), and face recognition (Guillamet & Vitria, 2002).”