Filters








11,818 Hits in 7.0 sec

Deep Creations: Intellectual Property and the Automata

Jean-Marc Deltorn
2017 Frontiers in Digital Humanities  
By the same token, the dependence on preexisting, protected, artworks lays the ground for potential zones of friction with the rights holders of the source data that helped shape the generative model.  ...  The rapid progress of deep neural network architectures is allowing both to automate the production of artworks and to extend the domain of creative expression.  ...  The generation of musical creation may therefore require some additional cautionary procedures in the training of deep-generative models.  ... 
doi:10.3389/fdigh.2017.00003 fatcat:bsyxgjtbp5bhzdmkj2hpogj37q

Live Coding and Machine Learning is Dangerous: Show us your Algorithms

Shelly Knotts, Iván Paz
2022 Zenodo  
Machine learning encompasses computer algorithms able to learn through experience and by using data, with the primary aim of the optimization of automation processes.  ...  Although the authors do not advocate puritanism, in considering the combination of live coding and ML from the lens of the TOPLAP draft manifesto, we find an axis on which to generate discussion contemplating  ...  /2021/ implementation of algorithms focused on small data, sometimes as a response to the dependency of big data deep learning models on expensive hardware.  ... 
doi:10.5281/zenodo.5888356 fatcat:t44luekjqfgc3l5a42ru53es6m

RaveForce: A Deep Reinforcement Learning Environment for Music Generation

Qichao Lan, Jim Tørresen, Alexander Refsum Jensenius
2019 Proceedings of the SMC Conferences  
and Music Computing Network NordicSMC, project number 86892.  ...  Acknowledgments This work was partially supported by the Research Council of Norway through its Centres of Excellence scheme, project number 262762 and by NordForsks Nordic University Hub Nordic Sound  ...  As one of the pioneers in automated music generation, in the piece called Analogique A, Iannis Xenakis uses Markov models for the order of musical sections [13] .  ... 
doi:10.5281/zenodo.3249325 fatcat:5axw6jy5bje6bjue72oowubcdq

Al-terity: Non-Rigid Musical Instrument with Artificial Intelligence Applied to Real-Time Audio Synthesis

Koray Tahiroǧlu, Miranda Kastemaa, Oskar Koli
2020 Proceedings of the International Conference on New Interfaces for Musical Expression  
In this paper, we present the Al-terity, a non-rigid musical instrument that comprises a deep learning model with generative adversarial network architecture and use it for generating audio samples for  ...  The particular deep learning model we use for this instrument was trained with existing data set as input for purposes of further experimentation.  ...  of inputs and outputs using deep learning methods [10] , and for simulated humanrobot imitation of particular music performance [25] as well as for prediction of musical events to provide non-intrusive  ... 
doi:10.5281/zenodo.4813402 fatcat:ki5cybay6bbdnorqmk2mzffqmq

Directed Evolution in Live Coding Music Performance

Sandeep Dasari, Jason Freeman
2020 Zenodo  
The paper also details some key design decisions, implementation, and usage of a novel genetic algorithm API created for a popular live coding language.  ...  Traditional evolutionary applications in music focused on novelty search to create new sounds, sequences of notes or chords, and effects.  ...  Secondly, the lack of active control in the generation i.e. deciphering the deep learning parameters, the complex mapping of the hidden layers of a neural network makes it impractical for live performance  ... 
doi:10.5281/zenodo.4285412 fatcat:ewalnm7gifbq7fl662famswr5y

Data-Driven Generative Live Coding for Music Creation

Gordan Kreković, Antonio Pošćić
2021 Zenodo  
With the existing assortment of techniques for algorithmic composition and text generation, one of the challenges was selecting the most appropriate approach for the task.  ...  The notion of automating the process of live coding entails interesting philosophical, conceptual, and technical questions.  ...  One of the requirements on the generative system was to produce consistent sonic results that reflect musical ideas of the composer.  ... 
doi:10.5281/zenodo.5137907 fatcat:drnv6ym3wradtij6r6wubwl5by

Feature learning and deep architectures: new directions for music informatics

Eric J. Humphrey, Juan P. Bello, Yann LeCun
2013 Journal of Intelligent Information Systems  
Acknowledging breakthroughs in other perceptual AI domains, we offer that deep learning holds the potential to overcome each of these obstacles.  ...  Through conceptual arguments for feature learning and deeper processing architectures, we demonstrate how deep processing models are more powerful extensions of current methods, and why now is the time  ...  Potential Impact In addition to hopefully advancing the discipline beyond current glass ceilings, there are several potential benefits to the adoption and research of deep learning in music informatics  ... 
doi:10.1007/s10844-013-0248-5 fatcat:m6n4kfas7nbtvk6cja6hbgzihq

Pervasive Intelligence

Sebastian Vehlken
2018 Digital Culture & Society  
As is shown on the basis of some generative examples from the field of UAS, robot swarms are imagined to literally penetrate space and control it.  ...  This article seeks to situate collective or swarm robotics (SR) on a conceptual pane which on the one hand sheds light on the peculiar form of AI which is at play in such systems, whilst on the other hand  ...  (Hauptmann/Neidich 2010) Against this background, the special issue Rethinking AI explores and critically reflects the hype of neuroinformatics in AI discourses and the potential and limits of critique  ... 
doi:10.14361/dcs-2018-0108 fatcat:rkuidqpnfjehpaychdcs3ccure

Playing with machines: Using machine learning to understand automated copyright enforcement at scale

Joanne E Gray, Nicolas P Suzor
2020 Big Data & Society  
This article presents the results of methodological experimentation that utilises machine learning to investigate automated copyright enforcement on YouTube.  ...  We hope that this work provides a methodological base for continued experimentation with the use of digital and computational methods to enable large-scale analysis of the operation of automated systems  ...  Acknowledgements We thank Rosalie Gillett for outstanding research assistance.  ... 
doi:10.1177/2053951720919963 fatcat:f2khh3yjdzenhkgmdnfwn26ysa

Data2Vis: Automatic Generation of Data Visualizations Using Sequence to Sequence Recurrent Neural Networks [article]

Victor Dibia, Çağatay Demiralp
2018 arXiv   pre-print
Data2Vis generates visualizations that are comparable to manually-created visualizations in a fraction of the time, with potential to learn more complex visualization strategies at scale.  ...  Rapidly creating effective visualizations using expressive grammars is challenging for users who have limited time and limited skills in statistics and data visualization.  ...  Adopting a learning approach to designing automated visualization systems holds potential for improving the maintenance and scalability of such systems.  ... 
arXiv:1804.03126v3 fatcat:p3konqc6pvdivexi3rbs64vnje

DeepDrummer : Generating Drum Loops using Deep Learning and a Human in the Loop

Guillaume Alain, Maxime Chevalier-Boisvert, Frederic Osterrath, Remi Piche-Taillefer
2020 Zenodo  
We train a deep neural network classifier on audio data and show how it can be used as the core component of a system that generates drum loops.  ...  DeepDrummer is a drum loop generation tool that uses active learning to learn the preferences (or current artistic intentions) of a human user from a small number of interactions.  ...  Acknowledgments The authors would like to thank Devine Lu Linvega for many enlightening conversations, whether about the world of tools for writing music or the universe in general.  ... 
doi:10.5281/zenodo.4285361 fatcat:zoeismp5jfdddpp7iicnjjdoyu

Introduction: Rethinking AI. Neural Networks, Biometrics and the New Artificial Intelligence

Mathias Fuchs, Ramón Reichert, Mediarep, Philipps Universität Marburg
2020 Digital Culture & Society  
(Hauptmann/Neidich 2010) Against this background, the special issue Rethinking AI explores and critically reflects the hype of neuroinformatics in AI discourses and the potential and limits of critique  ...  Introduction for example, the systematic recognition and processing of patterns (texts, images) which are to be controlled by means of deep learning.  ... 
doi:10.25969/mediarep/13522 fatcat:gvro5uk76jf5rjgkq6d77h3pmm

Conditioning Deep Generative Raw Audio Models for Structured Automatic Music

Rachel Manzelli, Vijay Thakkar, Ali Siahkamari, Brian Kulis
2018 Zenodo  
Existing automatic music generation approaches that feature deep learning can be broadly classified into two types: raw audio models and symbolic models.  ...  We consider a Long Short Term Memory network to learn the melodic structure of different styles of music, and then use the unique symbolic generations from this model as a conditioning input to a WaveNet-based  ...  Symbolic Audio Models Most deep learning approaches for automatic music generation are based on symbolic representations of the music.  ... 
doi:10.5281/zenodo.1492374 fatcat:q6a75dmhxndqrkswkxdmctoq6i

Conditioning Deep Generative Raw Audio Models for Structured Automatic Music [article]

Rachel Manzelli, Vijay Thakkar, Ali Siahkamari, Brian Kulis
2018 arXiv   pre-print
Existing automatic music generation approaches that feature deep learning can be broadly classified into two types: raw audio models and symbolic models.  ...  We consider a Long Short Term Memory network to learn the melodic structure of different styles of music, and then use the unique symbolic generations from this model as a conditioning input to a WaveNet-based  ...  Symbolic Audio Models Most deep learning approaches for automatic music generation are based on symbolic representations of the music.  ... 
arXiv:1806.09905v1 fatcat:eqq64xppffdgdkuurqezpnjfcy

DDSP études for Tenor Saxophone & Violin

Guilherme Coelho
2021 Zenodo  
Fiebrink, and Google's Magenta research work and development of machine learning models for music.  ...  performances in the practice of AI interfaces for musical expression, machine musicianship and humanmachine interactions.  ...  Interactions with the DDSP model can open new reflections of artistic practices, performance contexts and music practices for new explorations of music to emerge.  ... 
doi:10.5281/zenodo.5137969 fatcat:5vgb4wamh5azllgunilgqou2oa
« Previous Showing results 1 — 15 out of 11,818 results