Neural Audio Synthesis of Musical Notes with WaveNet autoencoders

From statwiki
Revision as of 01:25, 22 March 2018 by Milievsk (talk | contribs) (Created page with "= Introduction = = Contributions = = Models = center == WaveNet Autoencoder == Mu-law encoding: == Baseline: Spectral Autoencoder...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Introduction

Contributions

Models

paper26-figure1-models.png

WaveNet Autoencoder

Mu-law encoding:


Baseline: Spectral Autoencoder

Training

The NSynth Dataset

Evaluation

Reconstruction

paper27-figure2-reconstruction.png

left

Interpolation in Timbre and Dynamics

paper27-figure3-interpolation.png

Entanglement of Pitch and Timbre

paper27-figure4-entanglement.png

Future Directions