Unsupervised Blind Source Separation with Variational Auto-Encoders - Télécom Paris Access content directly
Conference Papers Year :

Unsupervised Blind Source Separation with Variational Auto-Encoders

Abstract

Supervised source separation requires expensive synthetic datasets containing clean, ground truth-source signals, while unsupervised separation requires only data mixtures. Existing unsupervised methods still use supervision to avoid over-separation and compete with fully supervised methods. We present a new method of completely unsupervised single-channel blind source separation, based on variational auto-encoding, that automatically learns the correct number of sources in data mixtures and quantitatively outperforms the existing methods. A deep inference network disentangles (separates) data mixtures into low-dimensional latent source variables. A deep generative network individually decodes each latent source into its source signal, such that their sum represents the given mixture. Qualitative and quantitative results from separation experiments on pairs of randomly mixed MNIST handwritten digits and mixed audio spectrograms demonstrate that our method outperforms stateof-the-art unsupervised and semi-supervised methods, showing promise as a solution to this long-standing problem in computer vision and audition.
Fichier principal
Vignette du fichier
2021_eusipco_vae_camera_ready.pdf (1.4 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03255341 , version 1 (17-05-2022)

Identifiers

  • HAL Id : hal-03255341 , version 1

Cite

Julian Neri, Roland Badeau, Philippe Depalle. Unsupervised Blind Source Separation with Variational Auto-Encoders. 29th European Signal Processing Conference (EUSIPCO 2021), Aug 2021, Dublin, Ireland. ⟨hal-03255341⟩
122 View
225 Download

Share

Gmail Facebook Twitter LinkedIn More