Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 26
Encontrada(s) 1 página(s)

Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Predictability of music descriptor time series and its application to cover song detection

  • Serrà Julià, Joan
  • Kantz, Holger
  • Serra, Xavier
  • Andrzejak, Ralph Gregor
Intuitively, music has both predictable and unpredictable components. In this work we assess this qualitative statement in a quantitative way using common time series models fitted to state-of-the-art music descriptors. These descriptors cover different musical facets and are extracted from a large collection of real audio recordings comprising a variety of musical genres. Our findings show that music descriptor time series exhibit a certain predictability not only for short time intervals, but also for mid-term and relatively long intervals. This fact is observed independently of the descriptor, musical facet and time series model we consider. Moreover, we show that our findings are not only of theoretical relevance /nbut can also have practical impact. To this end we demonstrate that music predictability at relatively long time intervals can be exploited in a real-world application, namely the automatic identification of cover songs (i.e. different renditions or versions of the same musical piece). Importantly, this prediction strategy yields a parameter-free approach for cover song identification that is substantially faster, allows for reduced computational storage and still maintains highly competitive accuracies when compared to state-of-the-art systems., This work has been partially funded by the Deutscher Akademischer Austausch Dienst (A/09/96235), by the projects Classical Planet (MITYC: TSI-070100- 2009-407) and DRIMS (MICINN: TIN2009-14247-C02-01), by the Spanish Ministry of Education /nand Science (BFU2007-61710) and by the Max Planck Institute for the Physics of Complex Systems.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Characterization and exploitation of community structure in cover song networks

  • Serrà Julià, Joan
  • Zanin, Massimiliano
  • Herrera Boyer, Perfecto, 1964-
  • Serra, Xavier
The use of community detection algorithms is explored within the framework of cover song identification, i.e. the automatic detection of different audio renditions of the same underlying musical piece. Until now, this task has been posed as a typical query-by-example task, where one submits a query song and the system retrieves a list of possible matches ranked by their similarity to the query. In this work, we propose a new approach which uses song communities (clusters, groups) to provide more relevant answers to a given query. Starting from the output of a state-of-the-art system, songs are embedded in a complex weighted network whose links represent similarity (related musical content). Communities inside the network are then recognized as groups of covers and this information is used to enhance the results of the system. In particular, we show that this approach increases both the coherence and the accuracy of the system. Furthermore, we provide insight into the internal organization of individual cover song communities, showing that there is a tendency for the original song to be central within the community. We postulate that the methods and results presented here could be relevant to other query-by-example tasks., This work has been supported by the following projects: Classical Planet (TSI-070100–2009-407; MITYC) and DRIMS (TIN2009–14247-C02–01; MICINN).
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Folksonomy-based tag recommendation for collaborative tagging systems

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Collaborative tagging has emerged as a common solution for labelling and organising online digital content. However, collaborative tagging systems typically suffer from a number of issues such as tag scarcity or ambiguous labelling. As a result, the organisation and browsing of tagged content is far from being optimal. In this work the authors present a general scheme for building a folksonomy-based tag recommendation system to help users tagging online content resources. Based on this general scheme, the authorse describe eight tag recommendation methods and extensively evaluate them with data coming from two real-world large-scale datasets of tagged images and sound clips. Their results show that the proposed methods can effectively recommend relevant tags, given a set of input tags and tag co-occurrence information. Moreover, the authors show how novel strategies for selecting the appropriate number of tags to be recommended can significantly improve methods performances. Approaches such as the one presented here can be useful to obtain more comprehensive and coherent descriptions of tagged resources, thus allowing a better organisation, browsing and reuse of online content. Moreover, they can increase the value of folksonomies as reliable sources for knowledge-mining., This work is partially supported under BES2010-037309 FPI grant from the Spanish Ministry of Science and Innovation for the TIN2009-14247-C02-01 DRIMS project. JS acknowledges 2009-SGR-1434 from Generalitat de Catalunya, JAEDOC069/2010 from Consejo Superior de Investigaciones Cientificas, and FP7-ICT-2011-8-318770 from the European Commission.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Audio clip classification using social tags and the effect of tag expansion

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Comunicació presentada a la 53rd International Conference: Semantic audio, celebrada els dies 27 a 29 de gener de 2014 a Londres, Regne Unit., Methods for automatic sound and music classification are of great value when trying to organise the large amounts of unstructured, user-contributed audio content uploaded to online sharing platforms. Currently, most of these methods are based on the audio signal, leaving the exploitation of users’ annotations or other contextual data rather unexplored. In this paper, we describe a method for the automatic classification of audio clips which is solely based on user-supplied tags. As a novelty, the method includes a tag expansion step for increasing classification accuracy when audio clips are scarcely tagged. Our results suggest that very high accuracies can be achieved in tag-based audio classification (even for poorly or badly annotated clips), and that the proposed tag expansion step can, in some cases, significantly increase classification performance. We are interested in the use of the described classification method as a first step for tailoring assistive tagging systems to the particularities of different audio categories, and as a way to improve the overall quality of online user annotations., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation (TIN2009-14247-C02-01; F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from Consejo Superior de Investigaciones Cient´ıficas (J.S.), ICT-2011-8-318770 from the European Commission (J.S.), and FP7-2007-2013 / ERC grant agreement 267583 (CompMusic).
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Automatic performer identification in commercial monophonic Jazz performances

  • Ramírez, Rafael, 1966-
  • Maestre Gómez, Esteban
  • Serra, Xavier
We present a pattern recognition approach to the task of identifying performers from their interpretative styles. We investigate how professional musicians express their view of the musical content of musical pieces and how to use this information in order to automatically identify performers. We apply sound analysis techniques based on spectral models for extracting deviation patterns of parameters such as pitch, timing, amplitude and timbre characterising both the internal structure of notes and the musical context in which they appear. We describe successful performer identification case studies involving monophonic audio recordings of both score-guided and commercial improvised performances., This work was supported by the Spanish Ministry of Science and Innovation under grants TIN2006-14932-C02-01 ProSeMus Project and TIN2009-14247-C02-01 DRIMS Project.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

How much metadata do we need in music recommendation?: a subjective evaluation using preference sets

  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
Comunicació presentada a: 2th International Society for Music Information Retrieval Conference (ISMIR 2011) celebrat del 24 al 28 d'octubre de 2011 a Miami, EEUU., In this work we consider distance-based approaches to music recommendation, relying on an explicit set of music tracks provided by the user as evidence of his/her music preferences. Firstly, we propose a purely content-based approach, working on low-level (timbral, temporal, and tonal) and inferred high-level semantic descriptions of music. Secondly, we consider its simple refinement by adding a minimum amount of genre metadata. We compare the proposed approaches with one content-based and three metadata-based baselines. As such, we consider content-based approach working on inferred semantic descriptors, a tag-based recommender exploiting artist tags, a commercial black-box recommender partially employing collaborative filtering information, and a simple genre-based random recommender. We conduct a listening experiment with 19 participants. The obtained results reveal that although the low-level/semantic content-based approach does not achieve the performance of the baseline working exclusively on the inferred semantic descriptors, the proposed refinement provides significant improvement in the listeners’ satisfaction comparable with metadata-based approaches, and surpasses these approaches by the number of novel relevant recommendations. We conclude that the proposed content-based approach refined by simple genre metadata is suited for music discovery not only in the long-tail but also within popular music items., This research has been partially funded by the FI Grant of Generalitat de Catalunya (AGAUR) and the Buscamedia (CEN-20091026), Classical Planet (TSI-070100-2009-407, MITYC), and DRIMS (TIN2009-14247- C02-01, MICINN) projects.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Zipf’s law in short-time timbral codings of speech, music, and environmental sound signals

  • Haro Berois, Martín
  • Serrà Julià, Joan
  • Herrera Boyer, Perfecto, 1964-
  • Corral, Álvaro
Timbre is a key perceptual feature that allows discrimination between different sounds. Timbral sensations are highly dependent on the temporal evolution of the power spectrum of an audio signal. In order to quantitatively characterize such sensations, the shape of the power spectrum has to be encoded in a way that preserves certain physical and perceptual properties. Therefore, it is common practice to encode short-time power spectra using psychoacoustical frequency scales. In this paper, we study and characterize the statistical properties of such encodings, here called timbral code-words. In particular, we report on rank-frequency distributions of timbral code-words extracted from 740 hours of audio coming from disparate sources such as speech, music, and environmental sounds. Analogously to text corpora, we find a heavy-tailed Zipfian distribution with exponent close to one. Importantly, this distribution is found independently of different encoding decisions and regardless of the audio source. Further analysis on the intrinsic characteristics of most and least frequent code-words reveals that the most frequent code-words tend to have a more homogeneous structure. We also find that speech and music databases have specific, distinctive code-words while, in the case of the environmental sounds, this database-specific code-words are not present. Finally, we find that a Yule-Simon process with memory provides a reasonable quantitative approximation for our data, suggesting the existence of a common simple generative mechanism for all considered sound sources., Funding was received from Classical Planet: TSI-070100- 2009-407 (MITYC), www.mityc.es; DRIMS: TIN2009-14247-C02-01 (MICINN), www.micinn.es; FIS2009-09508, www.micinn.es; and 2009SGR-164, www.gencat.cat. JS acknowledges funding from Consejo Superior de Investigaciones Científicas (JAEDOC069/2010), www.csic.es; and Generalitat de Catalunya (2009-SGR-1434), www.gencat.cat.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Analysis of the impact of a tag recommendation system in a real-world folksonomy

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Collaborative tagging systems have emerged as a successful solution for annotating contributed resources to online sharing platforms, facilitating searching, browsing, and organizing their contents. To aid users in the annotation process, several tag recommendation methods have been proposed. It has been repeatedly hypothesized that these methods should contribute to improving annotation quality and reducing the cost of the annotation process. It has been also hypothesized that these methods should contribute to the consolidation of the vocabulary of collaborative tagging systems. However, to date, no empirical and quantitative result supports these hypotheses. In this work, we deeply analyze the impact of a tag recommendation system in the folksonomy of Freesound, a real-world and large-scale online sound sharing platform. Our results suggest that tag recommendation effectively increases vocabulary sharing among users of the platform. In addition, tag recommendation is shown to contribute to the convergence of the vocabulary as well as to a partial increase in the quality of annotations. However, according to our analysis, the cost of the annotation process does not seem to be effectively reduced. Our work is relevant to increase our understanding about the nature of tag recommendation systems and points to future directions for the further development of those systems and their analysis., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation, TIN2009-14247-C02-01 from the Spanish government (F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from CSIC (J.S.), FSE2007-2013 E. U. Social funds (J.S.), and FP7-2007-2013/ERC grant agreement 267583 (CompMusic; F.F. and X.S.).
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Taking advantage of editorial metadata to recommend music

  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
Comunicació presentada al 9th International Symposium on Computer Music Modeling and Retrieval, celebrat del 19 al 22 de juny de 2012 a Londres, Regne Unit., In this work we propose a novel approach to music recommendation based exclusively on editorial metadata. To this end, we propose to use a public database of music releases Discogs.com, which contains extensive information about artists, their releases and record labels. We rely on an explicit set of music tracks provided by the user as evidence of his/her music preferences to construct a user profile suitable for distance-based music recommendation. We evaluate the proposed method against two purely metadata-based approaches and one approach partially based on audio content in a listening experiment with 27 participants. The results of subjective evaluation show that the proposed method is competitive to the state-of-the-art recommenders based on commercial metadata, while being easily implemented relying only on open public data., The authors would like to thank all participants involved in the evaluation. This research has been partially supported by the FI Grant of Generalitat de Catalunya (AGAUR) and the Classical Planet (TSI-070100- 2009-407, MITYC), DRIMS (TIN2009-14247-C02-01, MICINN), and MIRES (EC-FP7 ICT-2011.1.5 Networked Media and Search Systems, grant agreement No. 287711) projects.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Extending sound sample descriptions through the extraction of community knowledge

  • Font Corbera, Frederic
  • Serra, Xavier
Comunicació presentada a la 19th International Conference (UMAP) que va tenir lloc de l'11 al 15 de juliol a Girona., Sound and music online services driven by communities of users are filled with large amounts of user-created content that has to be properly described. In these services, typical sound and music modeling is performed using either content-based or context-based strategies, but no special emphasis is given to the extraction of knowledge from the community. We outline a research plan in the context of Freesound.org and propose ideas about how audio clip sharing sites could adapt and take advantage of particular user communities to improve the descriptions of their content., This work is partially supported under BES-2010-037309 FPI grant from the Spanish Ministry of Science and Innovation for the TIN2009- 14247-C02-01 DRIMS project.
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Class-based tag recommendation and user-based evaluation in online audio clip sharing

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Online sharing platforms often rely on collaborative tagging systems for annotating content. In this way, users themselves annotate and describe the shared contents using textual labels, commonly called tags. These annotations typically suffer from a number of issues such as tag scarcity or ambiguous labelling. Hence, to minimise some of these issues, tag recommendation systems can be employed to suggest potentially relevant tags during the annotation process. In this work, we present a tag recommendation system and evaluate it in the context of an online platform for audio clip sharing. By exploiting domain-specific knowledge, the system we present is able to classify an audio clip among a number of predefined audio classes and to produce specific tag recommendations for the different classes. We perform an in-depth user-based evaluation of the recommendation method along with two baselines and a former version that we described in previous work. This user-based evaluation is further complemented with a prediction-based evaluation following standard information retrieval methodologies. Results show that the proposed tag recommendation method brings a statistically significant improvement over the previous method and the baselines. In addition, we report a number of findings based on the detailed analysis of user feedback provided during the evaluation process. The considered methods, when applied to real-world collaborative tagging systems, should serve the purpose of consolidating the tagging vocabulary and improving the quality of content annotations., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation (TIN2009-14247-C02-01; F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from CSIC (J.S.), ICT-2011-8-318770 from the European Commission (J.S.), and FP7-2007-2013/ERC Grant Agreement 267583 (CompMusic; F.F., X.S.).
Repositorio Digital de la UPF
Publicación Artículo científico (JournalArticle)

Multi-feature beat tracking

  • Zapata González, José Ricardo
  • Davies, Matthew E. P.
  • Gómez Gutiérrez, Emilia, 1975-
A recent trend in the field of beat tracking for musical audio signals has been to explore techniques for measuring the level of agreement and disagreement between a committee of beat tracking algorithms. By using beat tracking evaluation methods to compare all pairwise combinations of beat tracker outputs, it has been shown that selecting the beat tracker which most agrees with the remainder of the committee, on a song-by-song basis, leads to improved performance which surpasses the accuracy of any individual beat tracker used on its own. In this paper we extend this idea towards presenting a single, standalone beat tracking solution which can exploit the benefit of mutual agreement without the need to run multiple separate beat tracking algorithms. In contrast to existing work, we re-cast the problem as one of selecting between the beat outputs resulting from a single beat tracking model with multiple, diverse input features. Through extended evaluation on a large annotated database, we show that our multi-feature beat tracker can outperform the state of the art, and thereby demonstrate that there is sufficient diversity in input features for beat tracking, without the need for multiple tracking models., This work was supported in part by the R+I+D Ph.D. scholarship of Universidad Pontificia Bolivariana and Colciencias (Colombia), the projects of the Spanish Ministry of Science and Innovation DRIMS (MICINN-TIN2009-14247-C02-01), SIGMUS (MINECO-TIN2012-36650) and Mires (EC-7PM-MIReS), and in part by the Media Arts and Technologies project (MAT), NORTE-07-0124-FEDER-000061, financed by the North Portugal Regional Operational Programme (ON.2 O Novo Norte), under the National Strategic Reference Framework (NSRF), through the European Regional Development Fund (ERDF) by national funds, through the Portuguese funding agency, Fundação para a Ciência e a Tecnologia (FCT), the FCT post-doctoral grant (SFRH/BPD/88722/2012), the PHENICX (EU: FP7 2007/2013 grant agreement n 601166) and COFLA (Junta de Andaluca P09-TIC-4840) projects.
Repositorio Digital de la UPF
Publicación Contribución de congreso (ConferenceOutput)

Music retagging using label propagation and robust principal component analysis

  • Yang, Yi-Hsuan
  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
  • Sordo, Mohamed
Comunicació presentada a: WWW'12 Companion: The 21st International Conference on World Wide Web celebrat del 16 al 20 d'abril de 2012 a Lió, França., The emergence of social tagging websites such as Last.fm has provided new opportunities for learning computational models that automatically tag music. Researchers typically obtain music tags from the Internet and use them to construct machine learning models. Nevertheless, such tags are usually noisy and sparse. In this paper, we present a preliminary study that aims at refining (retagging) social tags by exploiting the content similarity between tracks and the semantic redundancy of the track-tag matrix. The evaluated algorithms include a graph-based label propagation method that is often used in semi-supervised learning and a robust principal component analysis (PCA) algorithm that has led to state-of-the-art results in matrix completion. The results indicate that robust PCA with content similarity constraint is particularly effective; it improves the robustness of tagging against three types of synthetic errors and boosts the recall rate of music auto-tagging by 7% in a real-world setting., This work was supported by a grant from the National Science Council of Taiwan under contract NSC 100-2218-E001-009 and partially supported by the following projects: Classical Planet: TSI-070100-2009-407 (MITYC), DRIMS: TIN2009-14247-C02-01 (MICINN) and MIRES: EC-FP7 ICT2011.1.5 Networked Media and Search Systems, grant agreement No. 287711.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Analysis of the impact of a tag recommendation system in a real-world folksonomy

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Collaborative tagging systems have emerged as a successful solution for annotating contributed resources to online sharing platforms, facilitating searching, browsing, and organizing their contents. To aid users in the annotation process, several tag recommendation methods have been proposed. It has been repeatedly hypothesized that these methods should contribute to improving annotation quality and reducing the cost of the annotation process. It has been also hypothesized that these methods should contribute to the consolidation of the vocabulary of collaborative tagging systems. However, to date, no empirical and quantitative result supports these hypotheses. In this work, we deeply analyze the impact of a tag recommendation system in the folksonomy of Freesound, a real-world and large-scale online sound sharing platform. Our results suggest that tag recommendation effectively increases vocabulary sharing among users of the platform. In addition, tag recommendation is shown to contribute to the convergence of the vocabulary as well as to a partial increase in the quality of annotations. However, according to our analysis, the cost of the annotation process does not seem to be effectively reduced. Our work is relevant to increase our understanding about the nature of tag recommendation systems and points to future directions for the further development of those systems and their analysis., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation, TIN2009-14247-C02-01 from the Spanish government (F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from CSIC (J.S.), FSE2007-2013 E. U. Social funds (J.S.), and FP7-2007-2013/ERC grant agreement 267583 (CompMusic; F.F. and X.S.).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Music retagging using label propagation and robust principal component analysis

  • Yang, Yi-Hsuan
  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
  • Sordo, Mohamed
Comunicació presentada a: WWW'12 Companion: The 21st International Conference on World Wide Web celebrat del 16 al 20 d'abril de 2012 a Lió, França., The emergence of social tagging websites such as Last.fm has provided new opportunities for learning computational models that automatically tag music. Researchers typically obtain music tags from the Internet and use them to construct machine learning models. Nevertheless, such tags are usually noisy and sparse. In this paper, we present a preliminary study that aims at refining (retagging) social tags by exploiting the content similarity between tracks and the semantic redundancy of the track-tag matrix. The evaluated algorithms include a graph-based label propagation method that is often used in semi-supervised learning and a robust principal component analysis (PCA) algorithm that has led to state-of-the-art results in matrix completion. The results indicate that robust PCA with content similarity constraint is particularly effective; it improves the robustness of tagging against three types of synthetic errors and boosts the recall rate of music auto-tagging by 7% in a real-world setting., This work was supported by a grant from the National Science Council of Taiwan under contract NSC 100-2218-E001-009 and partially supported by the following projects: Classical Planet: TSI-070100-2009-407 (MITYC), DRIMS: TIN2009-14247-C02-01 (MICINN) and MIRES: EC-FP7 ICT2011.1.5 Networked Media and Search Systems, grant agreement No. 287711.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Taking advantage of editorial metadata to recommend music

  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
Comunicació presentada al 9th International Symposium on Computer Music Modeling and Retrieval, celebrat del 19 al 22 de juny de 2012 a Londres, Regne Unit., In this work we propose a novel approach to music recommendation based exclusively on editorial metadata. To this end, we propose to use a public database of music releases Discogs.com, which contains extensive information about artists, their releases and record labels. We rely on an explicit set of music tracks provided by the user as evidence of his/her music preferences to construct a user profile suitable for distance-based music recommendation. We evaluate the proposed method against two purely metadata-based approaches and one approach partially based on audio content in a listening experiment with 27 participants. The results of subjective evaluation show that the proposed method is competitive to the state-of-the-art recommenders based on commercial metadata, while being easily implemented relying only on open public data., The authors would like to thank all participants involved in the evaluation. This research has been partially supported by the FI Grant of Generalitat de Catalunya (AGAUR) and the Classical Planet (TSI-070100- 2009-407, MITYC), DRIMS (TIN2009-14247-C02-01, MICINN), and MIRES (EC-FP7 ICT-2011.1.5 Networked Media and Search Systems, grant agreement No. 287711) projects.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Characterization and exploitation of community structure in cover song networks

  • Serrà Julià, Joan
  • Zanin, Massimiliano
  • Herrera Boyer, Perfecto, 1964-
  • Serra, Xavier
The use of community detection algorithms is explored within the framework of cover song identification, i.e. the automatic detection of different audio renditions of the same underlying musical piece. Until now, this task has been posed as a typical query-by-example task, where one submits a query song and the system retrieves a list of possible matches ranked by their similarity to the query. In this work, we propose a new approach which uses song communities (clusters, groups) to provide more relevant answers to a given query. Starting from the output of a state-of-the-art system, songs are embedded in a complex weighted network whose links represent similarity (related musical content). Communities inside the network are then recognized as groups of covers and this information is used to enhance the results of the system. In particular, we show that this approach increases both the coherence and the accuracy of the system. Furthermore, we provide insight into the internal organization of individual cover song communities, showing that there is a tendency for the original song to be central within the community. We postulate that the methods and results presented here could be relevant to other query-by-example tasks., This work has been supported by the following projects: Classical Planet (TSI-070100–2009-407; MITYC) and DRIMS (TIN2009–14247-C02–01; MICINN).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Multi-feature beat tracking

  • Zapata González, José Ricardo
  • Davies, Matthew E. P.
  • Gómez Gutiérrez, Emilia, 1975-
A recent trend in the field of beat tracking for musical audio signals has been to explore techniques for measuring the level of agreement and disagreement between a committee of beat tracking algorithms. By using beat tracking evaluation methods to compare all pairwise combinations of beat tracker outputs, it has been shown that selecting the beat tracker which most agrees with the remainder of the committee, on a song-by-song basis, leads to improved performance which surpasses the accuracy of any individual beat tracker used on its own. In this paper we extend this idea towards presenting a single, standalone beat tracking solution which can exploit the benefit of mutual agreement without the need to run multiple separate beat tracking algorithms. In contrast to existing work, we re-cast the problem as one of selecting between the beat outputs resulting from a single beat tracking model with multiple, diverse input features. Through extended evaluation on a large annotated database, we show that our multi-feature beat tracker can outperform the state of the art, and thereby demonstrate that there is sufficient diversity in input features for beat tracking, without the need for multiple tracking models., This work was supported in part by the R+I+D Ph.D. scholarship of Universidad Pontificia Bolivariana and Colciencias (Colombia), the projects of the Spanish Ministry of Science and Innovation DRIMS (MICINN-TIN2009-14247-C02-01), SIGMUS (MINECO-TIN2012-36650) and Mires (EC-7PM-MIReS), and in part by the Media Arts and Technologies project (MAT), NORTE-07-0124-FEDER-000061, financed by the North Portugal Regional Operational Programme (ON.2 O Novo Norte), under the National Strategic Reference Framework (NSRF), through the European Regional Development Fund (ERDF) by national funds, through the Portuguese funding agency, Fundação para a Ciência e a Tecnologia (FCT), the FCT post-doctoral grant (SFRH/BPD/88722/2012), the PHENICX (EU: FP7 2007/2013 grant agreement n 601166) and COFLA (Junta de Andaluca P09-TIC-4840) projects.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Audio clip classification using social tags and the effect of tag expansion

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Comunicació presentada a la 53rd International Conference: Semantic audio, celebrada els dies 27 a 29 de gener de 2014 a Londres, Regne Unit., Methods for automatic sound and music classification are of great value when trying to organise the large amounts of unstructured, user-contributed audio content uploaded to online sharing platforms. Currently, most of these methods are based on the audio signal, leaving the exploitation of users’ annotations or other contextual data rather unexplored. In this paper, we describe a method for the automatic classification of audio clips which is solely based on user-supplied tags. As a novelty, the method includes a tag expansion step for increasing classification accuracy when audio clips are scarcely tagged. Our results suggest that very high accuracies can be achieved in tag-based audio classification (even for poorly or badly annotated clips), and that the proposed tag expansion step can, in some cases, significantly increase classification performance. We are interested in the use of the described classification method as a first step for tailoring assistive tagging systems to the particularities of different audio categories, and as a way to improve the overall quality of online user annotations., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation (TIN2009-14247-C02-01; F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from Consejo Superior de Investigaciones Cient´ıficas (J.S.), ICT-2011-8-318770 from the European Commission (J.S.), and FP7-2007-2013 / ERC grant agreement 267583 (CompMusic).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Automatic performer identification in commercial monophonic Jazz performances

  • Ramírez, Rafael, 1966-
  • Maestre Gómez, Esteban
  • Serra, Xavier
We present a pattern recognition approach to the task of identifying performers from their interpretative styles. We investigate how professional musicians express their view of the musical content of musical pieces and how to use this information in order to automatically identify performers. We apply sound analysis techniques based on spectral models for extracting deviation patterns of parameters such as pitch, timing, amplitude and timbre characterising both the internal structure of notes and the musical context in which they appear. We describe successful performer identification case studies involving monophonic audio recordings of both score-guided and commercial improvised performances., This work was supported by the Spanish Ministry of Science and Innovation under grants TIN2006-14932-C02-01 ProSeMus Project and TIN2009-14247-C02-01 DRIMS Project.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Extending sound sample descriptions through the extraction of community knowledge

  • Font Corbera, Frederic
  • Serra, Xavier
Comunicació presentada a la 19th International Conference (UMAP) que va tenir lloc de l'11 al 15 de juliol a Girona., Sound and music online services driven by communities of users are filled with large amounts of user-created content that has to be properly described. In these services, typical sound and music modeling is performed using either content-based or context-based strategies, but no special emphasis is given to the extraction of knowledge from the community. We outline a research plan in the context of Freesound.org and propose ideas about how audio clip sharing sites could adapt and take advantage of particular user communities to improve the descriptions of their content., This work is partially supported under BES-2010-037309 FPI grant from the Spanish Ministry of Science and Innovation for the TIN2009- 14247-C02-01 DRIMS project.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Class-based tag recommendation and user-based evaluation in online audio clip sharing

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Online sharing platforms often rely on collaborative tagging systems for annotating content. In this way, users themselves annotate and describe the shared contents using textual labels, commonly called tags. These annotations typically suffer from a number of issues such as tag scarcity or ambiguous labelling. Hence, to minimise some of these issues, tag recommendation systems can be employed to suggest potentially relevant tags during the annotation process. In this work, we present a tag recommendation system and evaluate it in the context of an online platform for audio clip sharing. By exploiting domain-specific knowledge, the system we present is able to classify an audio clip among a number of predefined audio classes and to produce specific tag recommendations for the different classes. We perform an in-depth user-based evaluation of the recommendation method along with two baselines and a former version that we described in previous work. This user-based evaluation is further complemented with a prediction-based evaluation following standard information retrieval methodologies. Results show that the proposed tag recommendation method brings a statistically significant improvement over the previous method and the baselines. In addition, we report a number of findings based on the detailed analysis of user feedback provided during the evaluation process. The considered methods, when applied to real-world collaborative tagging systems, should serve the purpose of consolidating the tagging vocabulary and improving the quality of content annotations., This work has been supported by BES-2010-037309 FPI from the Spanish Ministry of Science and Innovation (TIN2009-14247-C02-01; F.F.), 2009-SGR-1434 from Generalitat de Catalunya (J.S.), JAEDOC069/2010 from CSIC (J.S.), ICT-2011-8-318770 from the European Commission (J.S.), and FP7-2007-2013/ERC Grant Agreement 267583 (CompMusic; F.F., X.S.).
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Zipf’s law in short-time timbral codings of speech, music, and environmental sound signals

  • Haro Berois, Martín
  • Serrà Julià, Joan
  • Herrera Boyer, Perfecto, 1964-
  • Corral, Álvaro
Timbre is a key perceptual feature that allows discrimination between different sounds. Timbral sensations are highly dependent on the temporal evolution of the power spectrum of an audio signal. In order to quantitatively characterize such sensations, the shape of the power spectrum has to be encoded in a way that preserves certain physical and perceptual properties. Therefore, it is common practice to encode short-time power spectra using psychoacoustical frequency scales. In this paper, we study and characterize the statistical properties of such encodings, here called timbral code-words. In particular, we report on rank-frequency distributions of timbral code-words extracted from 740 hours of audio coming from disparate sources such as speech, music, and environmental sounds. Analogously to text corpora, we find a heavy-tailed Zipfian distribution with exponent close to one. Importantly, this distribution is found independently of different encoding decisions and regardless of the audio source. Further analysis on the intrinsic characteristics of most and least frequent code-words reveals that the most frequent code-words tend to have a more homogeneous structure. We also find that speech and music databases have specific, distinctive code-words while, in the case of the environmental sounds, this database-specific code-words are not present. Finally, we find that a Yule-Simon process with memory provides a reasonable quantitative approximation for our data, suggesting the existence of a common simple generative mechanism for all considered sound sources., Funding was received from Classical Planet: TSI-070100- 2009-407 (MITYC), www.mityc.es; DRIMS: TIN2009-14247-C02-01 (MICINN), www.micinn.es; FIS2009-09508, www.micinn.es; and 2009SGR-164, www.gencat.cat. JS acknowledges funding from Consejo Superior de Investigaciones Científicas (JAEDOC069/2010), www.csic.es; and Generalitat de Catalunya (2009-SGR-1434), www.gencat.cat.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

How much metadata do we need in music recommendation?: a subjective evaluation using preference sets

  • Bogdanov, Dmitry
  • Herrera Boyer, Perfecto, 1964-
Comunicació presentada a: 2th International Society for Music Information Retrieval Conference (ISMIR 2011) celebrat del 24 al 28 d'octubre de 2011 a Miami, EEUU., In this work we consider distance-based approaches to music recommendation, relying on an explicit set of music tracks provided by the user as evidence of his/her music preferences. Firstly, we propose a purely content-based approach, working on low-level (timbral, temporal, and tonal) and inferred high-level semantic descriptions of music. Secondly, we consider its simple refinement by adding a minimum amount of genre metadata. We compare the proposed approaches with one content-based and three metadata-based baselines. As such, we consider content-based approach working on inferred semantic descriptors, a tag-based recommender exploiting artist tags, a commercial black-box recommender partially employing collaborative filtering information, and a simple genre-based random recommender. We conduct a listening experiment with 19 participants. The obtained results reveal that although the low-level/semantic content-based approach does not achieve the performance of the baseline working exclusively on the inferred semantic descriptors, the proposed refinement provides significant improvement in the listeners’ satisfaction comparable with metadata-based approaches, and surpasses these approaches by the number of novel relevant recommendations. We conclude that the proposed content-based approach refined by simple genre metadata is suited for music discovery not only in the long-tail but also within popular music items., This research has been partially funded by the FI Grant of Generalitat de Catalunya (AGAUR) and the Buscamedia (CEN-20091026), Classical Planet (TSI-070100-2009-407, MITYC), and DRIMS (TIN2009-14247- C02-01, MICINN) projects.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Folksonomy-based tag recommendation for collaborative tagging systems

  • Font Corbera, Frederic
  • Serrà Julià, Joan
  • Serra, Xavier
Collaborative tagging has emerged as a common solution for labelling and organising online digital content. However, collaborative tagging systems typically suffer from a number of issues such as tag scarcity or ambiguous labelling. As a result, the organisation and browsing of tagged content is far from being optimal. In this work the authors present a general scheme for building a folksonomy-based tag recommendation system to help users tagging online content resources. Based on this general scheme, the authorse describe eight tag recommendation methods and extensively evaluate them with data coming from two real-world large-scale datasets of tagged images and sound clips. Their results show that the proposed methods can effectively recommend relevant tags, given a set of input tags and tag co-occurrence information. Moreover, the authors show how novel strategies for selecting the appropriate number of tags to be recommended can significantly improve methods performances. Approaches such as the one presented here can be useful to obtain more comprehensive and coherent descriptions of tagged resources, thus allowing a better organisation, browsing and reuse of online content. Moreover, they can increase the value of folksonomies as reliable sources for knowledge-mining., This work is partially supported under BES2010-037309 FPI grant from the Spanish Ministry of Science and Innovation for the TIN2009-14247-C02-01 DRIMS project. JS acknowledges 2009-SGR-1434 from Generalitat de Catalunya, JAEDOC069/2010 from Consejo Superior de Investigaciones Cientificas, and FP7-ICT-2011-8-318770 from the European Commission.
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Predictability of music descriptor time series and its application to cover song detection

  • Serrà Julià, Joan
  • Kantz, Holger
  • Serra, Xavier
  • Andrzejak, Ralph Gregor
Intuitively, music has both predictable and unpredictable components. In this work we assess this qualitative statement in a quantitative way using common time series models fitted to state-of-the-art music descriptors. These descriptors cover different musical facets and are extracted from a large collection of real audio recordings comprising a variety of musical genres. Our findings show that music descriptor time series exhibit a certain predictability not only for short time intervals, but also for mid-term and relatively long intervals. This fact is observed independently of the descriptor, musical facet and time series model we consider. Moreover, we show that our findings are not only of theoretical relevance /nbut can also have practical impact. To this end we demonstrate that music predictability at relatively long time intervals can be exploited in a real-world application, namely the automatic identification of cover songs (i.e. different renditions or versions of the same musical piece). Importantly, this prediction strategy yields a parameter-free approach for cover song identification that is substantially faster, allows for reduced computational storage and still maintains highly competitive accuracies when compared to state-of-the-art systems., This work has been partially funded by the Deutscher Akademischer Austausch Dienst (A/09/96235), by the projects Classical Planet (MITYC: TSI-070100- 2009-407) and DRIMS (MICINN: TIN2009-14247-C02-01), by the Spanish Ministry of Education /nand Science (BFU2007-61710) and by the Max Planck Institute for the Physics of Complex Systems.