Search Results for author: Gabriel Meseguer-Brocal

Found 8 papers, 7 papers with code

Detecting music deepfakes is easy but actually hard

1 code implementation7 May 2024 Darius Afchar, Gabriel Meseguer-Brocal, Romain Hennequin

In the face of a new era of generative models, the detection of artificially generated content has become a matter of utmost importance.

DeepFake Detection Face Swapping

An Experimental Comparison Of Multi-view Self-supervised Methods For Music Tagging

1 code implementation14 Apr 2024 Gabriel Meseguer-Brocal, Dorian Desblancs, Romain Hennequin

In this study, we expand the scope of pretext tasks applied to music by investigating and comparing the performance of new self-supervised methods for music tagging.

Contrastive Learning Music Tagging +1

A Lightweight Instrument-Agnostic Model for Polyphonic Note Transcription and Multipitch Estimation

1 code implementation18 Mar 2022 Rachel M. Bittner, Juan José Bosch, David Rubinstein, Gabriel Meseguer-Brocal, Sebastian Ewert

Despite its simplicity, benchmark results show our system's note estimation to be substantially better than a comparable baseline, and its frame-level accuracy to be only marginally below those of specialized state-of-the-art AMT systems.

Music Transcription

MULTIMODAL ANALYSIS: Informed content estimation and audio source separation

no code implementations27 Apr 2021 Gabriel Meseguer-Brocal

This dissertation proposes the study of multimodal learning in the context of musical signals.

Audio Source Separation

Data Cleansing with Contrastive Learning for Vocal Note Event Annotations

1 code implementation5 Aug 2020 Gabriel Meseguer-Brocal, Rachel Bittner, Simon Durand, Brian Brost

We propose a novel data cleansing model for time-varying, structured labels which exploits the local structure of the labels, and demonstrate its usefulness for vocal note event annotations in music.

Contrastive Learning Information Retrieval +2

Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source Separations

2 code implementations2 Jul 2019 Gabriel Meseguer-Brocal, Geoffroy Peeters

The input vector is embedded to obtain the parameters that control Feature-wise Linear Modulation (FiLM) layers.

Audio Source Separation

DALI: a large Dataset of synchronized Audio, LyrIcs and notes, automatically created using teacher-student machine learning paradigm

2 code implementations25 Jun 2019 Gabriel Meseguer-Brocal, Alice Cohen-Hadria, Geoffroy Peeters

We start with a set of manual annotations of draft time-aligned lyrics and notes made by non-expert users of Karaoke games.

Cannot find the paper you are looking for? You can Submit a new open access paper.