Self-Supervised VQ-VAE for One-Shot Music Style Transfer
arXiv:2102.05749 · doi:10.1109/ICASSP39728.2021.9414235
Abstract
Neural style transfer, allowing to apply the artistic style of one image to another, has become one of the most widely showcased computer vision applications shortly after its introduction. In contrast, related tasks in the music audio domain remained, until recently, largely untackled. While several style conversion methods tailored to musical signals have been proposed, most lack the 'one-shot' capability of classical image style transfer algorithms. On the other hand, the results of existing one-shot audio style transfer methods on musical inputs are not as compelling. In this work, we are specifically interested in the problem of one-shot timbre transfer. We present a novel method for this task, based on an extension of the vector-quantized variational autoencoder (VQ-VAE), along with a simple self-supervised learning strategy designed to obtain disentangled representations of timbre and pitch. We evaluate the method using a set of objective metrics and show that it is able to outperform selected baselines.
arXiv commentICASSP 2021. Website: https://adasp.telecom-paris.fr/s/ss-vq-vae
References in corpus (12)
- Adam: A Method for Stochastic Optimization
- Efficient Estimation of Word Representations in Vector Space
- WaveNet: A Generative Model for Raw Audio
- Neural Discrete Representation Learning
- Unsupervised Representation Learning by Predicting Image Rotations
- The challenge of realistic music generation: modelling raw audio at scale
- DDSP: Differentiable Digital Signal Processing
- Audio style transfer
- A Universal Music Translation Network
- Learning Interpretable Representation for Controllable Polyphonic Music Generation
- Assisted Sound Sample Generation with Musical Conditioning in Adversarial Auto-Encoders
- Vector-Quantized Timbre Representation