Unsupervised Deep Clustering for Source Separation: Direct Learning from Mixtures using Spatial Information

11/05/2018
by   Efthymios Tzinis, et al.
0

We present a monophonic source separation system that is trained by only observing mixtures with no ground truth separation information. We use a deep clustering approach which trains on multi-channel mixtures and learns to project spectrogram bins to source clusters that correlate with various spatial features. We show that using such a training process we can obtain separation performance that is as good as making use of ground truth separation information. Once trained, this system is capable of performing sound separation on monophonic inputs, despite having learned how to do so using multi-channel recordings.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/06/2018

Bootstrapping single-channel source separation via unsupervised spatial clustering on stereo mixtures

Separating an audio scene into isolated sources is a fundamental problem...
research
10/04/2019

Modeling the Comb Filter Effect and Interaural Coherence for Binaural Source Separation

Typical methods for binaural source separation consider only the direct ...
research
04/02/2019

Unsupervised training of a deep clustering model for multichannel blind source separation

We propose a training scheme to train neural network-based source separa...
research
10/23/2019

Bootstrapping deep music separation from primitive auditory grouping principles

Separating an audio scene such as a cocktail party into constituent, mea...
research
08/24/2018

Multi-scenario deep learning for multi-speaker source separation

Research in deep learning for multi-speaker source separation has receiv...
research
10/20/2021

REAL-M: Towards Speech Separation on Real Mixtures

In recent years, deep learning based source separation has achieved impr...
research
01/03/2021

Adversarial Unsupervised Domain Adaptation for Harmonic-Percussive Source Separation

This paper addresses the problem of domain adaptation for the task of mu...

Please sign up or login with your details

Forgot password? Click here to reset