Deep Music Analogy Via Latent Representation Disentanglement

06/09/2019
by   Ruihan Yang, et al.
8

Analogy is a key solution to automated music generation, featured by its ability to generate both natural and creative pieces based on only a few examples. In general, an analogy is made by partially transferring the music abstractions, i.e., high-level representations and their relationships, from one piece to another; however, this procedure requires disentangling music representations, which takes little effort for musicians but is non-trivial for computers. Three sub-problems arise: extracting latent representations from the observation, disentangling the representations so that each part has a unique semantic interpretation, and mapping the latent representations back to actual music. An explicitly-constrained conditional variational auto-encoder (EC2-VAE) is proposed as a unified solution to all three sub-problems. In this study, we focus on disentangling the pitch and rhythm representations of 8-beat music clips conditioned on chords. In producing music analogies, this model helps us to realize the imaginary situation of "what if" a piece is composed using a different pitch contour, rhythm pattern, chord progression etc., by borrowing the representations from other pieces. Finally, we validate the proposed disentanglement method using objective measurements and evaluate the analogy examples by a subjective study.

READ FULL TEXT

page 3

page 6

research
04/18/2019

Inspecting and Interacting with Meaningful Music Representations using VAE

Variational Autoencoders(VAEs) have already achieved great results on im...
research
11/26/2021

Learning source-aware representations of music in a discrete latent space

In recent years, neural network based methods have been proposed as a me...
research
06/21/2019

Classical Music Prediction and Composition by means of Variational Autoencoders

This paper proposes a new model for music prediction based on Variationa...
research
08/01/2021

Is Disentanglement enough? On Latent Representations for Controllable Music Generation

Improving controllability or the ability to manipulate one or more attri...
research
09/15/2022

Domain Adversarial Training on Conditional Variational Auto-Encoder for Controllable Music Generation

The variational auto-encoder has become a leading framework for symbolic...
research
10/27/2021

Exploring single-song autoencoding schemes for audio-based music structure analysis

The ability of deep neural networks to learn complex data relations and ...
research
06/06/2023

Emotion-Conditioned Melody Harmonization with Hierarchical Variational Autoencoder

Existing melody harmonization models have made great progress in improvi...

Please sign up or login with your details

Forgot password? Click here to reset