Filters








17 Hits in 1.6 sec

The AcousticBrainz Genre Dataset: Multi-Source, Multi-Level, Multi-Label, and Large-Scale

Dmitry Bogdanov, Alastair Porter, Hendrik Schreiber, Julián Urbano, Sergio Oramas
2019 Zenodo  
This paper introduces the AcousticBrainz Genre Dataset, a large-scale collection of hierarchical multi-label genre annotations from different metadata sources.  ...  Genre labels for the dataset are sourced from both expert annotations and crowds, permitting comparisons between strict hierarchies and folksonomies.  ...  We also thank tagtraum industries for providing the Tagtraum genre annotations.  ... 
doi:10.5281/zenodo.3527818 fatcat:xkc345fzgjduvblwkrv3dhbaxe

FMA: A Dataset For Music Analysis [article]

Michaël Defferrard, Kirell Benzi, Pierre Vandergheynst, Xavier Bresson
2017 arXiv   pre-print
The community's growing interest in feature and end-to-end learning is however restrained by the limited availability of large audio datasets.  ...  We here describe the dataset and how it was created, propose a train/validation/test split and three subsets, discuss some suitable MIR tasks, and evaluate some baselines for genre recognition.  ...  Looking at Table 1 , the well-known MagnaTagATune [20] and the Million Song Dataset (MSD) [3] as well as the newer AudioSet [10] and AcousticBrainz [32] appear as contenders for a large-scale  ... 
arXiv:1612.01840v3 fatcat:i7hmi4pp2rbsvii66xya3px3hm

Machine learning for music genre: multifaceted review and experimentation with audioset

Jaime Ramírez, M. Julia Flores
2019 Journal of Intelligent Information Systems  
Its main goal is give the reader an overview of the history and the current state-of-the-art, exploring techniques and datasets used to the date, as well as identifying current challenges, such as this  ...  Although research has been prolific in terms of number of published works, the topic still suffers from a problem in its foundations: there is no clear and formal definition of what genre is.  ...  Acknowledgements This work has been partially funded by FEDER funds and the Spanish Government (MICINN) through projects SBPLY/17/180501/000493 and TIN2016-77902-C3-1-P.  ... 
doi:10.1007/s10844-019-00582-9 fatcat:ajs4sfhtufd6lijtkf4icjhtii

Leveraging knowledge bases and parallel annotations for music genre translation

Elena Epure, Anis Khlif, Romain Hennequin
2019 Zenodo  
We call this a translation task and identify three cases: 1) no common annotated corpus between source and target tag systems exists, 2) such a large corpus exists, 3) only few common annotations exist  ...  Here, we choose a new angle for the genre study by seeking to predict what would be the genres of musical items in a target tag system, knowing the genres assigned to them within source tag systems.  ...  Dataset The dataset used in the experiments was created from the dataset used in the 2018 AcousticBrainz Genre Task, part of the MediaEval benchmarking initiative [7] .  ... 
doi:10.5281/zenodo.3527943 fatcat:x42s2sjjwzbeznrbdaj7jfk43u

Exploring Music Similarity With Acousticbrainz

Philip Tovstogan, Dmitry Bogdanov, Alastair Porter
2018 Zenodo  
Using user feedback as the evaluation source we are able to compare performance of various algorithms at large scale.  ...  In this thesis we address this problem and propose a system that can be used to perform subjective evaluation of different similarity metrics at large scale.  ...  Using user feedback as the evaluation source we are able to compare performance of various algorithms at large scale.  ... 
doi:10.5281/zenodo.1479768 fatcat:bmljiqt3tnh6zbfbhcehgvjx5m

TensorFlow Audio Models in Essentia [article]

Pablo Alonso-Jiménez, Dmitry Bogdanov, Jordi Pons, Xavier Serra
2020 arXiv   pre-print
Essentia is a reference open-source C++/Python library for audio and music analysis.  ...  In particular, we assess the generalization capabilities in a cross-collection evaluation utilizing both external tag datasets as well as manual annotations tailored to the taxonomies of our models.  ...  The resulting genre annotations are multi-label, and to evaluate each group of classifiers (corresponding to one of our in-house datasets) we use the subset of tracks that have a ground-truth label matching  ... 
arXiv:2003.07393v1 fatcat:4umpevhss5arxnma333v3sypdu

Knowledge Extraction And Representation Learning For Music Recommendation And Classification

Sergio Oramas, Xavier Serra
2017 Zenodo  
Next, we focus on learning new data representations from multimodal content using deep learning architectures, addressing the problems of cold-start music recommendation and multi-label music genre classification  ...  In this thesis, we address the problems of classifying and recommending music present in large collections.  ...  Multimodal dataset To the best of our knowledge, there are no publicly available large-scale datasets that encompass audio, images, text, and multi-label genre annotations.  ... 
doi:10.5281/zenodo.1048497 fatcat:kdh5jhvocbh3riwln6n2f756su

Knowledge Extraction And Representation Learning For Music Recommendation And Classification

Sergio Oramas, Xavier Serra
2017 Zenodo  
Next, we focus on learning new data representations from multimodal content using deep learning architectures, addressing the problems of cold-start music recommendation and multi-label music genre classification  ...  In this thesis, we address the problems of classifying and recommending music present in large collections.  ...  Multimodal dataset To the best of our knowledge, there are no publicly available large-scale datasets that encompass audio, images, text, and multi-label genre annotations.  ... 
doi:10.5281/zenodo.1100973 fatcat:yfpmc6qxbbakjp6qzvywyoaoci

Love Me, Love Me, Say (and Write!) that You Love Me: Enriching the WASABI Song Corpus with Lyrics Annotations [article]

Michael Fell, Elena Cabrio, Elmahdi Korfed, Michel Buffa, Fabien Gandon
2020 arXiv   pre-print
Such corpus labels and the provided methods can be exploited by music search engines and music professionals (e.g. journalists, radio presenters) to better handle large collections of lyrics, allowing  ...  We present the WASABI Song Corpus, a large corpus of songs enriched with metadata extracted from music databases on the Web, and resulting from the processing of song lyrics and from audio analysis.  ...  Acknowledgement This work is partly funded by the French Research National Agency (ANR) under the WASABI project (contract ANR-16-CE23-0017-01) and by the EU Horizon 2020 research and innovation programme  ... 
arXiv:1912.02477v2 fatcat:eltjjg2kjbcynkdtutgmfgyw2u

Music Classification: Beyond Supervised Learning, Towards Real-world Applications [article]

Minz Won, Janne Spijkervet, Keunwoo Choi
2021 Zenodo  
NOTE: We strongly recommend visiting https://music-classification.github.io/tutorial/ and use a web version of the book.  ...  In this book, we focus on the more modern history of music classification since the popularization of deep learning in mid 2010s.  ...  When the classification task has multiple labels, we need to aggregate multiple ROC-AUC scores and PR-AUC scores. In scikit-learn library, there is an option called average.  ... 
doi:10.5281/zenodo.5703780 fatcat:vpjixx4nmfaqtipf3ytuu7srwa

Music Classification: Beyond Supervised Learning, Towards Real-world Applications [article]

Minz Won, Janne Spijkervet, Keunwoo Choi
2021 Zenodo  
NOTE: We strongly recommend visiting https://music-classification.github.io/tutorial/ and use a web version of the book.  ...  In this book, we focus on the more modern history of music classification since the popularization of deep learning in mid 2010s.  ...  When the classification task has multiple labels, we need to aggregate multiple ROC-AUC scores and PR-AUC scores. In scikit-learn library, there is an option called average.  ... 
doi:10.5281/zenodo.5703779 fatcat:ggefiongcnb5boahjsz4lgiuz4

Improving Generalization of Deep Learning Music Classifiers

Morgan Buisson, Pablo Alonso, Dmitry Bogdanov
2021 Zenodo  
We also highlight the impact label noise can have in a small dataset setting and explore ways to improve the model's robustness.  ...  We first propose ways to maximize the amount of information extracted from small datasets through outliers detection and eÿcient audio data augmentation.  ...  Therefore, large quantities of labelled data are needed, which implies considerable efforts to retrieve and label substantial datasets.  ... 
doi:10.5281/zenodo.5554754 fatcat:thqdptf6qfcjtaz5txu5tmj6vq

Data Usage in MIR: History & Future Recommendations

Wenqin Chen, Jessica Keast, Jordan Moody, Corinne Moriarty, Felicia Villalobos, Virtue Winter, Xueqi Zhang, Xuanqi Lyu, Elizabeth Freeman, Jessie Wang, Sherry Cai, Katherine Kinnaird
2019 Zenodo  
from large companies and those within academia.  ...  As a result, there is an emerging divide in the MIR research community between labs that have access to music through large companies with abundant funds, and independent labs at smaller institutions who  ...  Datasets that used four or fewer genres were classified with each of the genres, those with five or more were given the genre label "various". 2 As many datasets contained multiple genres and many papers  ... 
doi:10.5281/zenodo.3527733 fatcat:jgwblxhu4ncg7cpvecdsmvcjgu

End-to-End Music Emotion Recognition: Towards Language-Sensitive Models

Ana G. Pandrea, Juan S. Gómez Cañón, Perfecto Herrera
2020 Zenodo  
The architecture is called SincNet and was initially proved to be successful for the task of speaker recognition.  ...  One problem could be that because each language has its particularities in terms of sound and intonation, and implicitly in terms of associations that are made upon them, we expect the observed emotions  ...  The extractor is suited for batch computations on large music collections and was used within AcousticBrainz [56] , a project that aims to crowd source acoustic information for all music in the world  ... 
doi:10.5281/zenodo.4091059 fatcat:kly3gq5hgbhilhd5vkfgfmjnii

The Semantic Web MIDI Tape

Albert Meroño-Peñuela, Reinier de Valk, Enrico Daga, Marilena Daquino, Anna Kent-Muller
2018 Proceedings of the 1st International Workshop on Semantic Applications for Audio and Music - SAAM '18  
The Linked Data paradigm has been used to publish a large number of musical datasets and ontologies on the Semantic Web, such as MusicBrainz, AcousticBrainz, and the Music Ontology.  ...  Despite the dataset making MIDI resources available in Web data standard formats such as RDF and SPARQL, the important issue of nding meaningful links between these MIDI resources and relevant contextual  ...  A GitHub organisation hosts all project repositories, including documentation and tutorials, source MIDI collections, and the dataset generation source code.  ... 
doi:10.1145/3243907.3243909 dblp:conf/semweb/Merono-PenuelaV18 fatcat:cqwwtgdxkneflpslwhvj2zehai
« Previous Showing results 1 — 15 out of 17 results