Multichannel audio source separation with deep neural networks

Aditya Arie Nugraha (1) , Antoine Liutkus (1) , Emmanuel Vincent (1)

1 MULTISPEECH - Speech Modeling for Facilitating Oral-Based Communication

This technical report considers the problem of multichannel audio source separation. A few studies have addressed the problem of single-channel audio source separation with deep neural networks (DNNs). We introduce a new framework for multichannel source separation where (1) spectral and spatial parameters are updated iteratively similarly to the expectation-maximization (EM) algorithm and (2) DNNs are used in the spectral updates. We evaluated several systems based on the proposed framework by participating in the "professionally-produced music recording" task of SiSEC 2015. Experimental results show that the framework performed well in separating singing voice and other instruments from a mixture containing multiple musical instruments.

Domaines: Traitement du signal et de l'image [eess.SP]

Aditya Arie Nugraha, Antoine Liutkus, Emmanuel Vincent. Multichannel audio source separation with deep neural networks. [Research Report] RR-8740, INRIA. 2015. ⟨hal-01163369v1⟩