Filters








2,393 Hits in 4.9 sec

Melody Generation for Pop Music via Word Representation of Musical Properties [article]

Andrew Shin, Leopold Crestel, Hiroharu Kato, Kuniaki Saito, Katsunori Ohnishi, Masataka Yamaguchi, Masahiro Nakawaki, Yoshitaka Ushiku, Tatsuya Harada
2017 arXiv   pre-print
Automatic melody generation for pop music has been a long-time aspiration for both AI researchers and musicians.  ...  Representation of multivariate property of notes has been one of the primary challenges.  ...  We generate melody with word representation of notes and their properties, instead of training multiple layers for each property, thereby reducing the complexity of learning.  ... 
arXiv:1710.11549v1 fatcat:poe6a3vq6bdgbklvozydi6w2g4

The Role of Features and Context in Recognition of Novel Melodies

Daniel Müllensiefen, Andrea R. Halpern
2014 Music Perception  
University of California Press is collaborating with JSTOR to digitize, preserve and extend access to Music Perception: An Interdisciplinary Journal.  ...  JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive.  ...  Author Note Correspondence concerning this article should be addressed to Daniel Mü llensiefen, Department of Psychology, Goldsmiths, University of London, New Cross Road, New Cross London SE14 6NW.  ... 
doi:10.1525/mp.2014.31.5.418 fatcat:djpptkormfczlgem4dqpbqjm4m

Melody2Vec: Distributed Representations of Melodic Phrases based on Melody Segmentation

Tatsunori Hirai, Shun Sawada
2019 Journal of Information Processing  
We assume phrases within melodies to be words and acquire these words via melody segmentation applying rules for grouping musical notes called Grouping Preference Rules (GPR) in the Generative Theory of  ...  We employed a skipgram representation to train our model using 10,853 melody tracks extracted from MIDI files primarily constructed from pop music.  ...  Acknowledgments We would like to thank Sakurako Yazawa for her advice on IRM symbol labeling and Yuko Kujime for her review of melody replacement result.  ... 
doi:10.2197/ipsjjip.27.278 fatcat:jjjau5tdq5g4tcmz3mqxo7f53a

The Effects of Musical and Linguistic Components in Recognition of Real-World Musical Excerpts by Cochlear Implant Recipients and Normal-Hearing Adults

K. Gfeller, D. Jiang, J. J. Oleson, V. Driscoll, C. Olszewski, J. F. Knutson, C. Turner, B. Gantz
2012 The Journal of music therapy  
Recognition by CI recipients improved as a function of linguistic cues. Participants were tested on melody recognition of complex melodies (pop, country, & classical styles).  ...  , lyric analysis, & listening for enjoyment).  ...  Acknowledgments This study was supported by research grant DC00242 from the National Institutes of Health/NIDCD; grant RR00059 from the General Clinical Research Centers Program, Division of Research Resources  ... 
doi:10.1093/jmt/49.1.68 pmid:22803258 pmcid:PMC3400117 fatcat:zc34bfssxnfgvaam5lrlcgrplm

Learning interpretable representation for controllable polyphonic music generation

Ziyu Wang, Dingsu Wang, Yixiao Zhang, Gus Xia
2020 Zenodo  
While deep generative models have become the leading methods for algorithmic composition, it remains a challenging problem to control the generation process because the latent variables of most deep-learning  ...  Both objective and subjective evaluations show that our method achieves a successful disentanglement and high quality controlled music generation.  ...  In this paper, we improve the model interpretability for music generation via constrained representation learning.  ... 
doi:10.5281/zenodo.4245518 fatcat:unc5fgfpevdjjbz2wi66lqp5oi

Learning Interpretable Representation for Controllable Polyphonic Music Generation [article]

Ziyu Wang, Dingsu Wang, Yixiao Zhang, Gus Xia
2020 arXiv   pre-print
While deep generative models have become the leading methods for algorithmic composition, it remains a challenging problem to control the generation process because the latent variables of most deep-learning  ...  Both objective and subjective evaluations show that our method achieves a successful disentanglement and high quality controlled music generation.  ...  In this paper, we improve the model interpretability for music generation via constrained representation learn- ing.  ... 
arXiv:2008.07122v1 fatcat:3lvtsv32jrgjbbcqhaqeqftdoa

MidiBERT-Piano: Large-scale Pre-training for Symbolic Music Understanding [article]

Yi-Hui Chou, I-Chun Chen, Chin-Jui Chang, Joann Ching, Yi-Hsuan Yang
2021 arXiv   pre-print
As such, our research can be taken as a benchmark for symbolic-domain music understanding.  ...  This paper presents an attempt to employ the mask language modeling approach of BERT to pre-train a 12-layer Transformer model over 4,166 pieces of polyphonic piano MIDI files for tackling a number of  ...  [36] , and Ching-Yu Chiu for sharing the Bi-LSTM code of her downbeat and beat tracker for audio [6] , which we adapted for implementing the baseline models.  ... 
arXiv:2107.05223v1 fatcat:5kyvtkwb55av3dqyxijduvosfi

Computational Creativity via Human-Level Concept Learning

Paul Bodily, Benjamin Bay, Dan Ventura
2017 International Conference on Computational Creativity  
, parsing, and generation of hand-written characters.  ...  We argue that the HBPL framework is wellsuited for modeling creative artefacts in general, one reason being that it allows explicit modeling of intention, structure, and substructure.  ...  of those in possession of large pop music datasets (sheet music sites, spotify, etc., asking for APIs, etc).  ... 
dblp:conf/icccrea/BodilyBV17 fatcat:ifgaygbsdrerjkfacvp2ld2gea

A Hierarchical Recurrent Neural Network for Symbolic Melody Generation [article]

Jian Wu and Changran Hu and Yulong Wang and Xiaolin Hu and Jun Zhu
2018 arXiv   pre-print
In this paper, we present a hierarchical recurrent neural network for melody generation, which consists of three Long-Short-Term-Memory (LSTM) subnetworks working in a coarse-to-fine manner along time.  ...  In recent years, neural networks have been used to generate symbolic melodies. However, the long-term structure in the melody has posed great difficulty for designing a good model.  ...  Acknowledgements This work was supported in part by the National Natural Science Foundation of China under Grant Nos. 61332007, 61621136008 and 61620106010.  ... 
arXiv:1712.05274v2 fatcat:vgrtwymrqbhxrnyyxfwrymxq24

Automatic Neural Lyrics and Melody Composition [article]

Gurunath Reddy Madhumani, Yi Yu, Florian Harscoët, Simon Canales, Suhua Tang
2020 arXiv   pre-print
AutoNLMC is designed to generate both lyrics and corresponding melody automatically for an amateur or a person without music knowledge.  ...  The qualitative and quantitative evaluation measures revealed that the proposed method is indeed capable of generating original lyrics and corresponding melody for composing new songs.  ...  An encoder-decoder based RNN sequential model for lyrics-conditional melody generation for Chinese pop songs is presented in [15] .  ... 
arXiv:2011.06380v1 fatcat:tsaogjgbuzg2jo5tocjreeux4m

Data-based melody generation through multi-objective evolutionary computation

Pedro J. Ponce de León, José M. Iñesta, Jorge Calvo-Zaragoza, David Rizo
2016 Journal of Mathematics and Music - Mathematical and Computational Approaches to Music Theory, Analysis, Composition and Performance  
Melodic trees are also proposed as a data structure for chromosomic representation of melodies and genetic operators are adapted to them.  ...  In this work, sets of melodies are utilized for training a machine learning approach to compute fitness, based on different metrics.  ...  Supplemental online material There is an Online Supplement where the reader can find some technical details, information about the data used, generated melodies, and additional information about the developed  ... 
doi:10.1080/17459737.2016.1188171 fatcat:qxctbnac6nap5o75g6yosgfj4i

The Cultural Significance of Timbre Analysis

Megan L. Lavengood
2020 Music Theory Online  
In Part 2, building from Allan Moore's definition of four functional layers in pop texture, I argue for the adoption of a fifth layer, which I term the novelty layer.  ...  This latter point is a reflection of the problematic treatment of world music by 1980s music culture.  ...  For my work in this article, I am taking pop music of 1950-1990 as my context.  ... 
doi:10.30535/mto.26.3.3 fatcat:r6wmtpoyunbuvelazapfoswudy

Language, music, and the brain: a resource-sharing framework [chapter]

Aniruddh D. Patel
2011 Language and Music as Cognitive Systems  
Supported by Neurosciences Research Foundation as part of its program on music and the brain at The Neurosciences Institute, where ADP is the Esther J. Burnham Senior Fellow. Notes  ...  Acknowledgements I thank John Iversen and Bob Slevc for helpful comments.  ...  For example, knowledge of words and their syntactic properties involves a set of representations which are distinct from the representations of chords and their harmonic relations.  ... 
doi:10.1093/acprof:oso/9780199553426.003.0022 fatcat:7feyuqgkg5ca5jsptqrlz2nlcq

Bowie musicology: mapping Bowie's sound and music language across the catalogue

Leah Kardos
2017 Continuum. Journal of Media and Cultural Studies  
the lyric and immediate pop/rock style representation.  ...  This music language uses vocal articulations, idiosyncratic approaches to melody and harmony, mode and tonality, familiar and foreign sonic landscapes and nostalgic references to encode meanings beyond  ...  Timbre, a word by which we refer to the quality or tone colour of a sound, is a property inextricably linked with communication and memory.  ... 
doi:10.1080/10304312.2017.1334387 fatcat:n5xbbxr6srcgvb34wp7lxxzd4e

Audio-Visual Content Analysis in P2P Networks: The SAPIR Approach

Walter Allasia, Fabrizio Falchi, Francesco Gallo, Mouna Kacimi, Aaron Kaplan, Jonathan Mamou, Yosi Mass, Nicola Orio
2008 2008 19th International Conference on Database and Expert Systems Applications  
The extracted features are then merged into a common representation. We report usage of this framework in the SAPIR demo.  ...  We present the SAPIR media framework for analyzing digital content and representing the extracted features in a common schema.  ...  We use the compact representation of word lattice called word confusion network (WCN) proposed by [5, 3] .  ... 
doi:10.1109/dexa.2008.123 dblp:conf/dexaw/AllasiaFGKKMMO08 fatcat:hjt4y2hvrbdvfciybigxqmby4u
« Previous Showing results 1 — 15 out of 2,393 results