Autoencoding Neural Networks as Musical Audio Synthesizers
Audio and Speech Processing
2020-04-29 v1 Machine Learning
Sound
Abstract
A method for musical audio synthesis using autoencoding neural networks is proposed. The autoencoder is trained to compress and reconstruct magnitude short-time Fourier transform frames. The autoencoder produces a spectrogram by activating its smallest hidden layer, and a phase response is calculated using real-time phase gradient heap integration. Taking an inverse short-time Fourier transform produces the audio signal. Our algorithm is light-weight when compared to current state-of-the-art audio-producing machine learning algorithms. We outline our design process, produce metrics, and detail an open-source Python implementation of our model.
Cite
@article{arxiv.2004.13172,
title = {Autoencoding Neural Networks as Musical Audio Synthesizers},
author = {Joseph Colonel and Christopher Curro and Sam Keene},
journal= {arXiv preprint arXiv:2004.13172},
year = {2020}
}