Denoising without access to clean data using a partitioned autoencoder

Abstract

Training a denoising autoencoder neural network requires access to truly clean data, a requirement which is often impractical. To remedy this, we introduce a method to train an autoencoder using only noisy data, having examples with and without the signal class of interest. The autoencoder learns a partitioned representation of signal and noise, learning to reconstruct each separately. We illustrate the method by denoising birdsong audio (available abundantly in uncontrolled noisy datasets) using a convolutional autoencoder.

References in corpus (6)

• ADADELTA: An Adaptive Learning Rate Method
• Theano: new features and speed improvements
• Exact solutions to the nonlinear dynamics of learning in deep linear neural networks
• Deep Convolutional Inverse Graphics Network
• Discovering Hidden Factors of Variation in Deep Networks
• Zero-bias autoencoders and the benefits of co-adapting features

Cited by in corpus (2)

• Feynman Machine: The Universal Dynamical Systems Computer
• Source separation with weakly labelled data: An approach to computational auditory scene analysis