publications
Sony CSL authors: Emmanuel Deruty, François Pachet, Pierre Roy
Abstract
The tremendous success of rock music in the second half of the 20th century has boosted the sophistication of production and mixing techniques for this music genre. However, there is no unified theory of mixing from the viewpoint of sound engineering. In this paper, we highlight relationships between loudness and spectrum in individual tracks, established during the process of mixing. To do so, we introduce an ad hoc, three-dimensional model of the spectrum of a track. These dimensions are derived from an optimal monitoring level, that is, the level that optimizes the number of frequency bands at the same, maximum loudness. We study a corpus of 55 rock multi-tracks and correlate the model with the loudness of the tracks. We suggest that (1) at high monitoring levels and/or on high-end monitors, track loudness is a linear function of its spectral centroid, and (2) at low monitoring levels and/or on budget monitors, a track’s optimal monitoring level is a linear function of its loudness. This indicates that under good listening conditions, human mixers tend to focus on spectral balance, whereas under bad conditions, they favor individual track comprehension. We discuss the implication of our results for automatic mixing.
Type: article
Keywords: audio, automatic mix
Downloads
BibTeX entry
@ARTICLE { pachet:14c, AUTHOR="Deruty, E. and Roy, P. and Pachet, F.", JOURNAL="Journal of the Audio Engineering Society", MONTH="October", NUMBER="10", PAGES="643--653", TITLE="Human-Made Rock Mixes Feature Tight Relations Between Spectrum And Loudness", VOLUME="62", YEAR="2014", }
Sony CSL authors: François Pachet
Abstract
Hit Song Science is an emerging field of investigation that aims at predicting the success of songs before they are released on the market. This chapter defines the context and goals of Hit Song Science from the viewpoint of music information retrieval. In the first part, we stress the complexity of the mechanisms underlying individual and social music preference from an experimental psychology viewpoint. In the second part, we describe current attempts at modeling and predicting music hits in a feature oriented view of popularity and, finally, draw conclusions on the current status of this emerging but fascinating field of research.
Type: incollection
Keywords: Hit Song Science, audio descriptors
Downloads
BibTeX entry
@INCOLLECTION { pachet:11a, AUTHOR="Pachet, F.", BOOKTITLE="Music Data Mining", CHAPTER="10", EDITOR="Tao, Tzanetakis & Ogihara", PAGES="305-326", PUBLISHER="Chapman & Hall / CRC Press", TITLE="Hit Song Science", YEAR="2011", }
Sony CSL authors: François Pachet, Pierre Roy
Abstract
We describe a large-scale experiment aiming at validating the hypothesis that the popularity of music titles can be predicted from global acoustic or human features. We use a 32.000 title database with 632 manually-entered labels per title including 3 related to the popularity of the title. Our experiment uses two audio feature sets, as well as the set of all the manually-entered labels but the popularity ones. The experiment shows that some subjective labels may indeed be reasonably well-learned by these techniques, but not popularity. This contradicts recent and sustained claims made in the MIR community and in the media about the existence of “Hit Song Science”.
Type: inproceedings
Keywords: feature extraction, timbre, Hit Song Science, audio descriptors
Downloads
BibTeX entry
@INPROCEEDINGS { pachet:08c, ADDRESS="Philadelphia, USA", AUTHOR="Pachet, F. and Roy, P.", BOOKTITLE="Proceedings of Ismir 2008", PAGES="355-360", TITLE="Hit Song Science is Not Yet a Science", YEAR="2008", }
Sony CSL authors: Jean-Julien Aucouturier, Anthony Beurivé, Amaury La Burthe, François Pachet, Aymeric Zils
Abstract
The IST project Cuidado, which ran from January 2001 to December 2003, produced the first entirely automatic chain for extracting and exploiting musical metadata for browsing music. The Sony CSL laboratory is primarily interested in the context of popular music browsing in large-scale catalogues. First, we are interested in human-centred issues related to browsing “Popular Music”. Popular here means that the music accessed to is widely distributed, and known to many listeners. Second, we consider “popular browsing” of music, i.e. making music accessible to non specialists (music lovers), and allowing sharing of musical tastes and information within communities, departing from the usual, single user view of digital libraries. This research project covers all areas of the music-to-listener chain, from music description - descriptor extraction from the music signal, or data mining techniques -, similarity based access and novel music retrieval methods such as automatic sequence generation, and user interface issues. This paper describes the scientific and technical issues at stake, and the results obtained.
Type: article
Keywords: Music Browser, Cuidado, audio descriptors
Downloads
BibTeX entry
@ARTICLE { pachet:04e, AUTHOR="Pachet, F., Aucouturier, J.-J., La Burthe, A., Zils, A. and Beurive, A.", JOURNAL="Multimedia Tools and Applications", NOTE="Special Issue on the CBMI03 Conference", NUMBER="3", PAGES="331-349", TITLE="The Cuidado Music Browser : an end-to-end Electronic Music Distribution System", VOLUME="30", YEAR="2006", }
Sony CSL authors: Atau Tanaka
Abstract
The new field of mobile music emerges at the intersection of ubiquitous computing, portable audio technology and NIME. We have held a series of international workshop on this topic with leading projects and speakers, in order to establish a community and stimulate the development of the field. In this report, we define mobile music, and map out the field by reporting on the workshop series and accounting for the state-of-the-art.
Type: inproceedings
Keywords: Mobile music, locative audio, mobile computing, ubiquitous computing, locative media, new interfaces for musical expression
Downloads
BibTeX entry
@INPROCEEDINGS { gaye:06a, AUTHOR="Gaye, L., Holmquist, L-E., Behrendt, F., Tanaka, A.", BOOKTITLE="Proceedings of New Interfaces for Musical Expression (NIME06)", ORGANIZATION="IRCAM", TITLE="Mobile Music Technology: Report on an Emerging Community", YEAR="2006", }
Sony CSL authors: Jean-Julien Aucouturier, Amaury La Burthe, François Pachet, Aymeric Zils
Abstract
The intentionally ambiguous expression "Popular Music Browser” reflects the two main goals of this project, which started in 1998, at Sony CSL laboratory. First, we are interested in human-centered issues related to browsing “Popular Music”. Popular here means that the music accessed to is widely distributed, and known to many listeners. Second, we consider “popular browsing” of music, i.e. making music accessible to non specialists (music lovers), and allowing sharing of musical tastes and information within communities, departing from the usual, single user view of digital libraries. This research project covers all areas of the music-to-listener chain, from music description - descriptor extraction from the music signal, or data mining techniques -, similarity based access and novel music retrieval methods such as automatic sequence generation, and user interface issues. This paper describes the scientific and technical issues at stake, and the results obtained, and is illustrated by prototypes developed within the European IST project Cuidado.
Type: article
Keywords: Cuidado, Music Browser, audio descriptors
Downloads
BibTeX entry
@ARTICLE { pachet:02a, AUTHOR="Pachet, F. Laburthe, A., Zils, A., Aucouturier, J.-J.", JOURNAL="Journal of the American Society for Information Science", NUMBER="12", PAGES="1037 - 1044", TITLE="Popular Music Access: The Sony Music Browser", VOLUME="55", YEAR="2004", }
Sony CSL authors: François Pachet, Aymeric Zils
Type: article
Keywords: EDS, feature generation, music, metadata, evolution, Cuidado, genetic programming, machine learning, audio descriptors
Downloads
BibTeX entry
@ARTICLE { pachet:03h, AUTHOR="Pachet, F. and Zils, A.", JOURNAL="Springer Verlag LNCS", PAGES="42-53", TITLE="Evolving Automatically High-Level Music Descriptors From Acoustic Signals", VOLUME="2771", YEAR="2003", }
Sony CSL authors: François Pachet, Aymeric Zils
Type: techreport
Keywords: Cuidado, metadata, rewriting, genetic programming, EDS, feature generation, audio descriptors
BibTeX entry
@TECHREPORT { pachet:03b, AUTHOR="Pachet F., Zils A.", INSTITUTION="Sony CSL", TITLE="Evolving Automatically High-Level Music Descriptors From Acoustic Signals", YEAR="2003", }
Sony CSL authors: François Pachet, Aymeric Zils
Abstract
We address the issue of extracting automatically high-level musical descriptors out of their raw audio signal. This work focuses on the extraction of the perceived intensity of music titles, that evaluates how energic the music is perceived by listeners. We present here first the perceptive tests that we have conducted, in order to evaluate the relevance and the universality of the perceived intensity descriptor. Then we present several methods used to extract relevant features used to build automatic intensity extractors: usual Mpeg7 low level features, empirical method, and features automatically found using our Extractor Discovery System (EDS), and compare the final performances of their extractors.
Type: inproceedings
Keywords: Cuidado, signal processing, audio descriptors, energy, genetic programming machine learning, feature generation
Downloads
BibTeX entry
@INPROCEEDINGS { zils:03a, ADDRESS="London, U.K.", AUTHOR="Zils, A. & Pachet, F.", BOOKTITLE="Proceedings of the 6th COST-G6 Conference on Digital Audio Effects (DAFX03)", MONTH="September", ORGANIZATION="Queen Mary University", PAGES="180-183", TITLE="Extracting Automatically the Perceived Intensity of Music Titles", YEAR="2003", }
Sony CSL authors: Jean-Julien Aucouturier, Amaury La Burthe, François Pachet
Abstract
The IST project Cuidado, which started in January 2001, aims at producing the first entirely automatic chain for extracting and exploiting musical metadata for browsing music. The Sony CSL laboratory is primarily interested in the context of popular music browsing in large-scale catalogues. First, we are interested in human-centered issues related to browsing “Popular Music”. Popular here means that the music accessed to is widely distributed, and known to many listeners. Second, we consider “popular browsing” of music, i.e. making music accessible to non specialists (music lovers), and allowing sharing of musical tastes and information within communities, departing from the usual, single user view of digital libraries. This research project covers all areas of the music-to-listener chain, from music description - descriptor extraction from the music signal, or data mining techniques -, similarity-based access and novel music retrieval methods such as automatic sequence generation, and user interface issues. This paper describes the scientific and technical issues at stake, and the results obtained, in the current state of the IST project.
Type: inproceedings
Keywords: Cuidado, Music Browser, audio descriptors
Downloads
BibTeX entry
@INPROCEEDINGS { pachet:03i, ADDRESS="Rennes, France", AUTHOR="Pachet, F. Laburthe, A., Aucouturier, J.-J.", BOOKTITLE="Proceedings of CBMI 03", EDITOR="INRIA", PAGES="27-34", PUBLISHER="IRISA", TITLE="The Cuidado Music Browser: an end-to-end electronic music distribution system", YEAR="2003", }
Sony CSL authors: Jean-Julien Aucouturier, Amaury La Burthe, François Pachet
Abstract
We address the metadata management problem in the context of future Electronic Music Distribution (EMD) systems. We propose a classification of existing musical editorial systems in two categories: the autistic and the universalists. Universalists propose shared information, at the expense of consensuality. Autistic approaches allow individual parameterization, at the expense of reusability. We propose an architecture and a system for managing editorial metadata that lies in the middle of these two extremes viewpoints: we organize musical editorial information in such a way that users can benefit from shared metadata when they wish, while allowing them to create and manage a private version of editorial information. A mechanism allows to synchronize both views (the shared and the private view). We describe the architecture and the application in progress, in the context of the Cuidado European IST project.
Type: inproceedings
Keywords: metadata, Cuidado, Music Browser, audio descriptors
Downloads
BibTeX entry
@INPROCEEDINGS { pachet:03e, ADDRESS="Leeds, U.K.", AUTHOR="La Burthe, A. and Pachet, F. and Aucouturier, JJ", BOOKTITLE="Proceedings of the WedelMusic 2003 Conference", EDITOR="Kia Ng. et al.", MONTH="September", PAGES="130-137", TITLE="Editorial Metadata in the Cuidado Music Browser: Between Universalism and Autism", YEAR="2003", }
Sony CSL authors: Olivier Delerue, François Pachet, Aymeric Zils
Abstract
An approach for extracting automatically time indexes of occurrences of percussive sounds in audio signals taken from the Popular music repertoire.
Type: inproceedings
Keywords: Cuidado, Audio Descriptors, Rhythm, Popular Music
Downloads
BibTeX entry
@INPROCEEDINGS { zils:02a, ADDRESS="Darmstadt, Germany", AUTHOR="Zils A., Pachet F., Delerue O., Gouyon F.", BOOKTITLE="Proceedings of WedelMusic 2002", MONTH="December", TITLE="Automatic Extraction of Drum Tracks from Polyphonic Music Signals", YEAR="2002", }
Sony CSL authors: François Pachet, Aymeric Zils
Abstract
This work addresses the issue of retrieving efficiently sound samples in large databases, in the context of digital music composition. We propose a sequence generation mechanism called musical mosaicing, which enables to generate automatically sequences of sound samples by specifying only high-level properties of the sequence to generate. The properties of the sequence specified by the user are translated automatically into constraints holding on descriptors of the samples. The system we propose is able to scale up on databases containing more than 100.000 samples, using a local search method based on constraint solving. In this paper, we describe the method for retrieving and sequencing audio samples, and illustrate it with rhythmic and melodic musical sequences.
Type: inproceedings
Keywords: musaicing, constraints, sequencing, audio descriptors
Downloads
BibTeX entry
@INPROCEEDINGS { zils:01a, ADDRESS="Limerick, U.K.", AUTHOR="Zils, A. and Pachet, F", BOOKTITLE="Proceedings of the COST G-6 Conference on Digital Audio Effects (DaFx-01)", MONTH="December", ORGANIZATION="University of Limerick", PAGES="39-44", TITLE="Musical Mosaicing", YEAR="2001", }
Sony CSL authors: Olivier Delerue, François Pachet
Type: inproceedings
Keywords: metadata, audio descriptors, rhythm
Downloads
BibTeX entry
@INPROCEEDINGS { gouyon:00a, ADDRESS="Verona, Italy", AUTHOR="Gouyon, F. and Delerue, O. and Pachet, F.", BOOKTITLE="Proceedings of DAFX 00", TITLE="Classifying percussive sounds: a matter of zero-crossing rate ?", YEAR="2000", }
Sony CSL authors: Olivier Delerue, François Pachet
Type: inproceedings
Keywords: audio descriptors, rhythm, Cuidado
BibTeX entry
@INPROCEEDINGS { pachet:00m, AUTHOR="Pachet, F. and Delerue, O. and Gouyon, F.", BOOKTITLE="Proceedings of SONY Research Forum", TITLE="Extracting Rhythm from Audio Signals", YEAR="2000", }