FSD50K: An Open Dataset of Human-Labeled Sound Events
- Eduardo Fonseca, Xavier Favory, Jordi Pons, F. Font, Xavier Serra
- 1 October 2020
Computer Science
IEEE/ACM Transactions on Audio Speech and…
FSD50K is introduced, an open dataset containing over 51 k audio clips totalling over 100 h of audio manually labeled using 200 classes drawn from the AudioSet Ontology, to provide an alternative benchmark dataset and thus foster SER research.
Freesound technical demo
- F. Font, Gerard Roma, Xavier Serra
- 21 October 2013
Computer Science
ACM Multimedia
This demo wants to introduce Freesound to the multimedia community and show its potential as a research resource.
Freesound Datasets: A Platform for the Creation of Open Audio Datasets
- Eduardo Fonseca, Jordi Pons, Xavier Serra
- 2017
Computer Science
International Society for Music Information…
Comunicacio presentada al 18th International Society for Music Information Retrieval Conference celebrada a Suzhou, Xina, del 23 al 27 d'cotubre de 2017.
General-purpose Tagging of Freesound Audio with AudioSet Labels: Task Description, Dataset, and Baseline
- Eduardo Fonseca, Manoj Plakal, Xavier Serra
- 26 July 2018
Computer Science
Workshop on Detection and Classification of…
The goal of the task is to build an audio tagging system that can recognize the category of an audio clip from a subset of 41 diverse categories drawn from the AudioSet Ontology.
Learning Sound Event Classifiers from Web Audio with Noisy Labels
- Eduardo Fonseca, Manoj Plakal, D. Ellis, F. Font, Xavier Favory, Xavier Serra
- 4 January 2019
Computer Science
IEEE International Conference on Acoustics…
Experiments suggest that training with large amounts of noisy data can outperform training with smaller amounts of carefully-labeled data, and it is shown that noise-robust loss functions can be effective in improving performance in presence of corrupted labels.
Audio tagging with noisy labels and minimal supervision
- Eduardo Fonseca, Manoj Plakal, F. Font, D. Ellis, Xavier Serra
- 7 June 2019
Computer Science
Workshop on Detection and Classification of…
This paper presents the task setup, the FSDKaggle2019 dataset prepared for this scientific evaluation, and a baseline system consisting of a convolutional neural network.
Freesound 2: An Improved Platform for Sharing Audio Clips
- V. Akkermans, F. Font, Xavier Serra
- 2011
Physics
Freesound.org is an online collaborative sound database where people from different disciplines share recorded sound clips under Creative Commons licenses. It was started in 2005 and it is being…
Audio Commons: bringing Creative Commons audio content to the creative industries
- F. Font, Tim S. Brookes, Xavier Serra
- 2 February 2016
Computer Science
The Audio Commons Initiative is presented, which is aimed at promoting the use of open audio content and at developing technologies with which to support the ecosystem composed by content repositories, production tools and users.
An Interpretable Deep Learning Model for Automatic Sound Classification
- Pablo Zinemanas, Martín Rocamora, M. Miron, F. Font, Xavier Serra
- 2 April 2021
Computer Science
Electronics
This work proposes a novel interpretable deep learning model for automatic sound classification, which explains its predictions based on the similarity of the input to a set of learned prototypes in a latent space by designing a frequency-dependent similarity measure and by considering different time-frequency resolutions in the feature space.
Class-based tag recommendation and user-based evaluation in online audio clip sharing
- F. Font, J. Serrà, Xavier Serra
- 1 September 2014
Computer Science
Knowledge-Based Systems
...
...