Deep neural networks for single channel source separation

By Emad M. Grais, M. U. Sen, Hakan Erdogan. 2013

139 citations recorded by the source corpus.

In this paper, a novel approach for single channel source separation (SCSS) using a deep neural network (DNN) architecture is introduced. Unlike previous studies in which DNN and other classifiers were used for classifying time-frequency bins to obtain hard masks for each source, we use the DNN to classify estimated source spectra to check for their validity during separation. In the training stage, the training data for the source signals are used to train a DNN. In the separation stage, the trained DNN is utilized to aid in estimation of each source in the mixed signal. Single channel source separation problem is formulated as an energy minimization problem where each source spectra estimate is encouraged to fit the trained DNN model and the mixed signal spectrum is encouraged to be written as a weighted sum of the estimated source spectra. The proposed approach works regardless of the energy scale differences between the source signals in the training and separation stages. Nonnegative matrix factorization (NMF) is used to initialize the DNN estimate for each source. The experimental results show that using DNN initialized by NMF for source separation improves the quality of the separated signal compared with using NMF for source separation.

Subfields: 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 Signal Processing

1. Dong Yu
2. Li Deng
3. Yuki Mitsufuji
4. Deliang Wang
5. Jinyu Li
6. Paris Smaragdis
7. Po-Sen Huang
8. Shinji Watanabe

Citations in the corpus

• Deep Learning: Methods and Applications (2014, background): “to emotion recognition from speech in [207, 222], to spoken language understanding in [242, 366, 403], to speaker recognition in [351, 372], to language-type recognition in [112], to dialogue state tracking for spoken dialogue systems in [94, 152], to automatic voice activity detection in [442], to ”
• End-To-End Source Separation With Adaptive Front-Ends (2017, background): “Common approaches include neural networks which given the noisy magnitudes, either predict a noiseless magnitude spectrum [5, 6, 7], or some type of a masking function [8, 9].”
• Deep neural network based instrument extraction from music (2015, result): “This is in contrast to other proposed DNN approaches for BSS of music which are supervised in the sense that they assume to have more knowledge about the signals [10] [11] [12].”
• Joint Optimization of Masks and Deep Recurrent Neural Networks for Monaural Source Separation (2015, methodology): “Since our goal is to separate one of the sources from a mixture, instead of learning one of the sources as the target, we adapt the framework from [27] to model all different sources simultaneously.”