Filters








1,254 Hits in 8.1 sec

Artificial Neural Networks Jamming on the Beat [article]

Alexey Tikhonov, Ivan P. Yamshchikov
2021 arXiv   pre-print
Exploring a latent space of drum patterns one could generate new drum patterns with a given music style.  ...  The paper presents a large dataset of drum patterns alongside with corresponding melodies. It explores two possible methods for drum pattern generation.  ...  One can sample an arbitrary point from the latent space and use the decoder to convert that point into a new pattern.  ... 
arXiv:2007.06284v2 fatcat:tjycsmob55at3phqwlkrrmtvjq

Representation Learning for Drum Loops with a Variational Autoencoder

James McDermott
2018 Zenodo  
A large library of MIDI drum loops is described and used to train an autoencoder neural network in an unsupervised fashion.  ...  The result is a low-dimension representation which captures essential dimensions of variation in the data, and can be used to generate new drum loops and interpolate between pairs of loops.  ...  Thanks to the anonymous reviewers for helpful suggestions.  ... 
doi:10.5281/zenodo.4285469 fatcat:v3n662fdezdfrjavwgzupu7zoy

PocketVAE: A Two-step Model for Groove Generation and Control [article]

Kyungyun Lee, Wonil Kim, Juhan Nam
2021 arXiv   pre-print
In order to model the drum notes, we apply a discrete latent representation method via Vector Quantized Variational Autoencoder (VQ-VAE), as drum notes have a discrete property, unlike velocity and microtiming  ...  Creating a good drum track to imitate a skilled performer in digital audio workstations (DAWs) can be a time-consuming process, especially for those unfamiliar with drums.  ...  Thus, we apply vector quantization to the latent encoding space to model drum notes, which brings significant performance improvement.  ... 
arXiv:2107.05009v1 fatcat:vepti7svxrgerdn37lecp2nm6q

Dualization Of Rhythm Patterns

Błażej Kotowski, Sergi Jordà, Behzad Haki
2020 Zenodo  
Then, we train a cluster of autoencoders to find a latent, two-dimensional representation of these h-vectors, which we treat as a dualized version of the input pattern.  ...  Rhythm pattern dualization is a transformation of a multi-instrumental rhythm pattern to another pattern composed of maximum two instruments while maintaining coherence and the perceptual essence of the  ...  To generate the output, all we have to do is to sample a regular latent vector z from the distribution, giving the parameters and pass it to the decoder.  ... 
doi:10.5281/zenodo.4091468 fatcat:ajcrlrjitrezzkaasjksswtnau

Magenta Studio: Augmenting Creativity with Deep Learning in Ableton Live

Adam Roberts, Jesse Engel, Yotam Mann, Jon Gillick, Claire Kayacik, Signe Nørly, Monica Dinculescu, Carey Radebaugh, Curtis Hawthorne, Douglas Eck
2019 Zenodo  
Combining compelling applications of deep learning-based music generation with a focus on ease of installation and use in a popular DAW, we hope to expose more musicians and producers to the potential  ...  We tie together existing frameworks (Electron, TensorFlow.js, and Max For Live) to develop a system whose purpose is to bring the promise of interactive MuMe to the realm of professional music creators  ...  Thanks to Colin Raffel for his thoughtful review of this paper. Thanks to the valuable feedback from early adopters and user study participants.  ... 
doi:10.5281/zenodo.4285265 fatcat:yjmxojcx4fbyngmofi3hhai6uu

A Benchmarking Initiative for Audio-Domain Music Generation Using the Freesound Loop Dataset [article]

Tun-Min Hung, Bo-Yu Chen, Yen-Tung Yeh, Yi-Hsuan Yang
2021 arXiv   pre-print
Moreover, we use a larger col-lection of drum loops from Looperman to establish fourmodel-based objective metrics for evaluation, releasingthese metrics as a library for quantifying and facilitatingthe  ...  This paper proposes a new benchmark task for generat-ing musical passages in the audio domain by using thedrum loops from the FreeSound Loop Dataset, which arepublicly re-distributable.  ...  ACKNOWLEDGEMENTS This research work is supported by the Ministry of Science and Technology (MOST), Taiwan, under grant number 109-2628-E-001-002-MY2.  ... 
arXiv:2108.01576v1 fatcat:suyf3sltsreofnc44ayyffb46i

A Minimal Template for Interactive Web-based Demonstrations of Musical Machine Learning [article]

Vibert Thio, Hao-Min Liu, Yin-Cheng Yeh, Yi-Hsuan Yang
2019 arXiv   pre-print
It also helps researchers to gain insights into different models and to A/B test them.  ...  The template comes with a small codebase, is open source, and is meant to be easy to use by any practitioners to implement their own demonstrations.  ...  The Latent Inspector, shown in Figure 2 , lets the user modify the latent vector of a drum pattern displayed in the browser to find out how the drum pattern will alter correspondingly.  ... 
arXiv:1902.03722v1 fatcat:olr3etzmejfmlffuly2txkep4e

A Benchmarking Initiative for Audio-domain Music Generation using the FreeSound Loop Dataset

Tun Min Hung, Bo-Yu Chen, Yen Tung Yeh, Yi-Hsuan Yang
2021 Zenodo  
Moreover, we use a larger collection of drum loops from Looperman to establish four model-based objective metrics for evaluation, releasing these metrics as a library for quantifying and facilitating the  ...  This paper proposes a new benchmark task for generating musical passages in the audio domain by using the drum loops from the FreeSound Loop Dataset, which are publicly re-distributable.  ...  This research work is supported by the Ministry of Science and Technology (MOST), Taiwan, under grant number 109-2628-E-001-002-MY2.  ... 
doi:10.5281/zenodo.5624409 fatcat:n7e5sdc4qzdclhsroa3ivmp7vi

BassNet: A Variational Gated Autoencoder for Conditional Generation of Bass Guitar Tracks with Learned Interactive Control

Maarten Grachten, Stefan Lattner, Emmanuel Deruty
2020 Applied Sciences  
We empirically show that the model can disentangle bass patterns that require sensitivity to harmony, instrument timbre, and rhythm.  ...  An innovative aspect of our work is that the model is trained to learn a temporally stable two-dimensional latent space variable that offers interactive user control.  ...  When the prior is known, it is possible to draw samples from the model, and to manually explore the latent space in order to better control the model output.  ... 
doi:10.3390/app10186627 doaj:763a95c03ae24d29b457c5c14cb7ce21 fatcat:zjbqnboi6bgadl46rmujc533em

Evolving building blocks of rhythm: How human cognition creates music via cultural transmission [article]

Andrea Ravignani, Bill Thompson, Thomas Grossi, Tania Delgado, Simon Kirby
2017 bioRxiv   pre-print
Data from two experiments is analyzed using two different, complementary techniques. In the experiments, participants hear drumming patterns and imitate them.  ...  These patterns are then given to the same or another participant to imitate. The structure of these - initially random - patterns is tracked down to later experimental 'generations'.  ...  Acknowledgments AR and BT are grateful to Marieke Woensdregt and Carmen Saldaña Gascón for hospitality and  ... 
doi:10.1101/198390 fatcat:vbkmrkirkvbefmndhv3btcuaom

Tap2Drum with Transformer Neural Networks

Marina S. Nieto Giménez, Sergi Jordà, Behzad Haki
2021 Zenodo  
In this task, the aim is to transform a tapped pattern into a full-fledged drum beat.  ...  In a paper recently published by the Google Magenta team, in the context of au-tomatic beat generation and beat humanization, a new task called Tap2Drum is described.  ...  , the objective is to generate a full drum beat from a very reduced abstraction: a tapped sequence.  ... 
doi:10.5281/zenodo.5554740 fatcat:xdp27lj57ngjnn76iht43gdfbi

Machine Learning for Computational Creativity: VST Synthesizer Programming

Christopher Mitcheltree, Hideki Koike
2021 Zenodo  
It is demonstrated that the system learns to prioritize effects and can discover more efficient effect order sequences than a variety of baselines.  ...  First, a system that provides step-by-step instructions for applying audio effects to change some input audio towards a desired sound is designed.  ...  Acknowledgements List of Acronyms Bibliography Acknowledgments I'd like to thank my advisor Professor Koike, the students and staff at the Koike Lab, everyone at Qosmo, and my friends and family for their  ... 
doi:10.5281/zenodo.6351291 fatcat:tqvuepndzjdnzbf2qv23ryweuq

Detecting tāla Computationally in Polyphonic Context-A Novel Approach

Susmita Bhaduri, Anirban Bhaduri, Dipak Ghosh
2018 American Journal of Computer Science and Information Technology  
We have exploited this cyclic property along with stressed and non-stressed methods of playing tabla-strokes to extract a characteristic pattern from the left-tabla strokes, which, after matching with  ...  The human auditory system uses perceptual grouping of musical-elements and easily filters the tabla component, thereby decoding prominent rhythmic features like tala, tempo from a polyphonic composition  ...  We also thank renowned musician This article is available from: https://www.imedpub.com/computer-science-and-information-technology/ Foundation) for extending help in editing the manuscript to enhance  ... 
doi:10.21767/2349-3917.100030 fatcat:nkbcr4uhoneslfy66niy37y264

Evolving building blocks of rhythm: how human cognition creates music via cultural transmission

Andrea Ravignani, Bill Thompson, Thomas Grossi, Tania Delgado, Simon Kirby
2018 Annals of the New York Academy of Sciences  
Latent variables Bayesian model: Sketch of how the algorithm processes two drumming patterns (workflow proceeds from top to bottom, and from left to right).  ...  n u s c r i p t Latent variables Bayesian model: Sketch of how the algorithm processes two drumming patterns (workflow proceeds from top to bottom, and from left to right).  ... 
doi:10.1111/nyas.13610 pmid:29508405 fatcat:bkwlipdcrbawfoyrf5edlauwd4

Towards Modeling And Decomposing Loop-Based Electronic Music

Patricio López-Serrano, Christian Dittmar, Jonathan Driedger, Meinard Müller
2016 Zenodo  
If we know a priori which patterns are included and wish to find their respective activation positions, we need a technique capable of identifying an audio query within a database where further musical  ...  By adding sparsity constraints and shift-invariant probabilistic latent component analysis (SI-PLCA), they automatically identify the number of patterns and their lengthsapplied to beat-synchronous chromagrams  ... 
doi:10.5281/zenodo.1417999 fatcat:hgc7hmguwzgcjlkg4pvqlewxw4
« Previous Showing results 1 — 15 out of 1,254 results