Unsupervised Deep Clustering for Source Separation: Direct Learning from Mixtures using Spatial Information
This addresses source separation for audio processing applications, offering an unsupervised approach that eliminates the need for labeled data.
The paper tackles monophonic source separation by training a deep clustering system on multi-channel mixtures without ground truth, achieving performance comparable to using ground truth information.
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.