Browsing by Author "Fonseca, Eduardo"

Sort by: Order: Results:

  • A Simple fusion of deep and shallow learning for acoustic scene classification
    Fonseca, Eduardo; Gong, Rong; Serra, Xavier (Cyprus University of Technology, 2018)
    In the past, Acoustic Scene Classification systems havebeen based on hand crafting audio features that are input toa classifier. Nowadays, the common trend is to adopt datadriven techniques, e.g., deep learning, where audio ...
  • Acoustic scene classification by ensembling gradient boosting machine and convolutional neural networks
    Fonseca, Eduardo; Gong, Rong; Bogdanov, Dmitry; Slizovskaia, Olga; Gómez Gutiérrez, Emilia, 1975-; Serra, Xavier (Tampere University of Technology, 2017)
    This work describes our contribution to the acoustic scene classifi- cation task of the DCASE 2017 challenge. We propose a system that consists of the ensemble of two methods of different nature: a feature engineering ...
  • Facilitating the manual annotation of sounds when using large taxonomies
    Favory, Xavier; Fonseca, Eduardo; Font Corbera, Frederic; Serra, Xavier (FRUCT, 2018)
    Properly annotated multimedia content is crucial for supporting advances in many Information Retrieval applications. It enables, for instance, the development of automatic tools for the annotation of large and diverse ...
  • Freesound datasets: a platform for the creation of open audio datasets
    Fonseca, Eduardo; Pons Puig, Jordi; Favory, Xavier; Font Corbera, Frederic; Bogdanov, Dmitry; Ferraro, Andrés; Oramas, Sergio; Porter, Alastair; Serra, Xavier (International Society for Music Information Retrieval (ISMIR), 2017)
    Openly available datasets are a key factor in the advancement of data-driven research approaches, including many of the ones used in sound and music computing. In the last few years, quite a number of new audio datasets ...
  • FSD50K: an open dataset of human-labeled sound events
    Fonseca, Eduardo; Favory, Xavier; Pons, Jordi; Font, Frederic; Serra, Xavier (Institute of Electrical and Electronics Engineers (IEEE), 2022)
    Most existing datasets for sound event recognition (SER) are relatively small and/or domain-specific, with the exception of AudioSet, based on over 2 M tracks from YouTube videos and encompassing over 500 sound classes. ...
  • General-purpose tagging of Freesound audio with audioSet labels: task description, dataset, and baseline
    Fonseca, Eduardo; Plakal, Manoj; Font Corbera, Frederic; Ellis, Daniel P. W.; Favory, Xavier; Pons Puig, Jordi; Serra, Xavier (Tampere University of Technology, 2018)
    This paper describes Task 2 of the DCASE 2018 Challenge, titled “General-purpose audio tagging of Freesound content with AudioSet labels”. This task was hosted on the Kaggle platform as “Freesound General-Purpose Audio ...
  • Genome-wide pathway analysis identifies genetic pathways associated with psoriasis
    Aterido Ballonga, Adrià, 1990-; Julià, Antonio; Ferrándiz, Carlos; Puig, Lluís; Fonseca, Eduardo; Fernández-López, Emilia; Dauden, Esteban; Sánchez-Carazo, José Luís; López-Estebaranz, José Luís; Moreno-Ramírez, David; Vanaclocha, Francisco; Herrera, Enrique; Cueva, Pablo de la; Dand, Nick; Palau, Núria; Alonso, Arnald; López-Lasanta, María; Tortosa, Raül; García-Montero, Andrés; Codó, Laia; Gelpí, Josep Lluís; Bertranpetit, Jaume, 1952-; Absher, Devin; Capon, Francesca; Myers, Richard M.; Barker, Jonathan N.; Marsal, Sara (Elsevier, 2016)
    Psoriasis is a chronic inflammatory disease with a complex genetic architecture. To date, the psoriasis heritability is only partially explained. However, there is increasing evidence that the missing heritability in ...
  • Learning sound event classifiers from web audio with noisy labels
    Fonseca, Eduardo; Plakal, Manoj; Ellis, Daniel P. W.; Font Corbera, Frederic; Favory, Xavier; Serra, Xavier (Institute of Electrical and Electronics Engineers (IEEE), 2018)
    As sound event classification moves towards larger datasets, issues of label noise become inevitable. Web sites can supply large volumes of user-contributed audio and metadata, but inferring labels from this metadata ...
  • Self-supervised learning from automatically separated sound scenes
    Fonseca, Eduardo; Jansen, Aren; Ellis, Daniel P. W.; Wisdom, Scott; Tagliasacchi, Marco; Hershey, John R.; Plakal, Manoj; Hershey, Shawn; Moore, R. Channing; Serra, Xavier (Institute of Electrical and Electronics Engineers (IEEE), 2021)
    Real-world sound scenes consist of time-varying collections of sound sources, each generating characteristic sound events that are mixed together in audio recordings. The association of these constituent sound events with ...
  • Toward interpretable polyphonic sound event detection with attention maps based on local prototypes
    Zinemanas, Pablo; Rocamora, Martín; Fonseca, Eduardo; Font, Frederic; Serra, Xavier (Universitat Pompeu Fabra. Music Technology Group, 2021)
    Understanding the reasons behind the predictions of deep neural networks is a pressing concern as it can be critical in several application scenarios. In this work, we present a novel interpretable model for polyphonic ...
  • Training sound event classifiers using different types of supervision
    Fonseca, Eduardo (Universitat Pompeu Fabra, 2021-12-01)
    The automatic recognition of sound events has gained attention in the past few years, motivated by emerging applications in fields such as healthcare, smart homes, or urban planning. When the work for this thesis started, ...
  • Unsupervised contrastive learning of sound event representations
    Fonseca, Eduardo; Ortego, Diego; McGuinness, Kevin; O’Connor, Noel E.; Serra, Xavier (Institute of Electrical and Electronics Engineers (IEEE), 2021)
    Self-supervised representation learning can mitigate the limitations in recognition tasks with few manually labeled data but abundant unlabeled data—a common scenario in sound event research. In this work, we explore ...