Demiurge

Roberto ALONSO TRILLO; Marek Poliks

Demiurge

Roberto ALONSO TRILLO (Developer)
, Marek Poliks (Developer)

Academy of Music

Research output: Non-textual form › Software

Abstract

Demiurge is a tripartite neural network architecture devised to generate and sequence audio waveforms. The architecture combines a synthesis engine based on a una-GAN + mel-GAN / hifi-GAN model with a custom transformer-based sequencer.

Original language	English
Media of output	Other
Publication status	Published - Dec 2020

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 17 Partnerships for the Goals

Access to Document

https://github.com/robertoalonsotrillo/_Demiurge_Licence: GNU GPL

Cite this

@misc{ea1b81c8e6634a41aec8c328e661a3f2,

title = "Demiurge",

abstract = "Demiurge is a tripartite neural network architecture devised to generate and sequence audio waveforms. The architecture combines a synthesis engine based on a una-GAN + mel-GAN / hifi-GAN model with a custom transformer-based sequencer.",

author = "\{ALONSO TRILLO\}, Roberto and Marek Poliks",

year = "2020",

month = dec,

language = "English",

}

TY - ADVS

T1 - Demiurge

A2 - ALONSO TRILLO, Roberto

A2 - Poliks, Marek

PY - 2020/12

Y1 - 2020/12

N2 - Demiurge is a tripartite neural network architecture devised to generate and sequence audio waveforms. The architecture combines a synthesis engine based on a una-GAN + mel-GAN / hifi-GAN model with a custom transformer-based sequencer.

AB - Demiurge is a tripartite neural network architecture devised to generate and sequence audio waveforms. The architecture combines a synthesis engine based on a una-GAN + mel-GAN / hifi-GAN model with a custom transformer-based sequencer.

M3 - Software

ER -

Demiurge

Abstract

UN SDGs

Access to Document

Fingerprint

Cite this