ATT: Attention-based Timbre Transfer

Jain, Deepak Kumar; Kumar, Akshi; Cai, Linqin; Singhal, Siddharth; Kumar, Vaibhav

doi:10.1109/ijcnn48605.2020.9207146

Cited by 4 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recent proliferation of deep neural networks as audio synthesisers is further expanding the capabilities of these tools: realistic instrument performances can be synthesised from simple, low dimensional control signals [1][2][3]; the timbre of one instrument can be convincingly transferred to another [1,[3][4][5]; instruments can be morphed and interpolated along nonlinear manifolds [6,7]; and sounds can be manipulated using high level descriptors of perceptual characteristics [7][8][9]. Yet despite their impressive abilities, these systems have not been widely adopted in music creation workflows.…”

Section: Introductionmentioning

confidence: 99%

Neural Waveshaping Synthesis

Hayes¹,

Saitis²,

Fazekas³

2021

Preprint

View full text Add to dashboard Cite

We present the Neural Waveshaping Unit (NEWT): a novel, lightweight, fully causal approach to neural audio synthesis which operates directly in the waveform domain, with an accompanying optimisation (FastNEWT) for efficient CPU inference. The NEWT uses time-distributed multilayer perceptrons with periodic activations to implicitly learn nonlinear transfer functions that encode the characteristics of a target timbre. Once trained, a NEWT can produce complex timbral evolutions by simple affine transformations of its input and output signals. We paired the NEWT with a differentiable noise synthesiser and reverb and found it capable of generating realistic musical instrument performances with only 260k total model parameters, conditioned on F0 and loudness features. We compared our method to state-of-the-art benchmarks with a multistimulus listening test and the Fréchet Audio Distance and found it performed competitively across the tested timbral domains. Our method significantly outperformed the benchmarks in terms of generation speed, and achieved real-time performance on a consumer CPU, both with and without FastNEWT, suggesting it is a viable basis for future creative sound design tools.

show abstract

Section: Introductionmentioning

confidence: 99%