Geoffroy Peeters

SSM-Net: feature learning for Music Structure Analysis using a Self-Similarity-Matrix based loss

Nov 15, 2022
Geoffroy Peeters, Florian Angulo

* Extended Abstracts for the Late-Breaking Demo Session of the 23rd Int. Society for Music Information Retrieval Conf., Bengaluru, India, 2022

Access Paper or Ask Questions

Exploiting Device and Audio Data to Tag Music with User-Aware Listening Contexts

Nov 14, 2022
Karim M. Ibrahim, Elena V. Epure, Geoffroy Peeters, Gaël Richard

Figure 2 for Exploiting Device and Audio Data to Tag Music with User-Aware Listening Contexts

Figure 3 for Exploiting Device and Audio Data to Tag Music with User-Aware Listening Contexts

Figure 4 for Exploiting Device and Audio Data to Tag Music with User-Aware Listening Contexts

* Published in ISMIR

Access Paper or Ask Questions

Deep-Learning Architectures for Multi-Pitch Estimation: Towards Reliable Evaluation

Feb 18, 2022
Christof Weiß, Geoffroy Peeters

Figure 2 for Deep-Learning Architectures for Multi-Pitch Estimation: Towards Reliable Evaluation

Figure 3 for Deep-Learning Architectures for Multi-Pitch Estimation: Towards Reliable Evaluation

Figure 4 for Deep-Learning Architectures for Multi-Pitch Estimation: Towards Reliable Evaluation

Access Paper or Ask Questions

Content based singing voice source separation via strong conditioning using aligned phonemes

Aug 05, 2020
Gabriel Meseguer-Brocal, Geoffroy Peeters

Figure 2 for Content based singing voice source separation via strong conditioning using aligned phonemes

Figure 3 for Content based singing voice source separation via strong conditioning using aligned phonemes

Figure 4 for Content based singing voice source separation via strong conditioning using aligned phonemes

* 21st International Society for Music Information Retrieval Conference 11-15 October 2020, Montreal, Canada

Access Paper or Ask Questions

Learning to rank music tracks using triplet loss

May 18, 2020
Laure Prétet, Gaël Richard, Geoffroy Peeters

Figure 2 for Learning to rank music tracks using triplet loss

Figure 3 for Learning to rank music tracks using triplet loss

Access Paper or Ask Questions

A Prototypical Triplet Loss for Cover Detection

Oct 22, 2019
Guillaume Doras, Geoffroy Peeters

Figure 2 for A Prototypical Triplet Loss for Cover Detection

Figure 3 for A Prototypical Triplet Loss for Cover Detection

Figure 4 for A Prototypical Triplet Loss for Cover Detection

Access Paper or Ask Questions

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

Jul 09, 2019
Gabriel Meseguer-Brocal, Geoffroy Peeters

Figure 2 for Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source Separations

Figure 3 for Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source Separations

Figure 4 for Conditioned-U-Net: Introducing a Control Mechanism in the U-Net for Multiple Source Separations

* Proceedings of the 20th International Society for Music Information Retrieval Conference, ISMIR, Delft, Netherlands, 2019

Access Paper or Ask Questions

Cover Detection using Dominant Melody Embeddings

Jul 03, 2019
Guillaume Doras, Geoffroy Peeters

Figure 2 for Cover Detection using Dominant Melody Embeddings

Figure 3 for Cover Detection using Dominant Melody Embeddings

Figure 4 for Cover Detection using Dominant Melody Embeddings

Access Paper or Ask Questions

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

Jun 25, 2019
Gabriel Meseguer-Brocal, Alice Cohen-Hadria, Geoffroy Peeters

Figure 1 for DALI: a large Dataset of synchronized Audio, LyrIcs and notes, automatically created using teacher-student machine learning paradigm

Figure 2 for DALI: a large Dataset of synchronized Audio, LyrIcs and notes, automatically created using teacher-student machine learning paradigm

Figure 3 for DALI: a large Dataset of synchronized Audio, LyrIcs and notes, automatically created using teacher-student machine learning paradigm

Figure 4 for DALI: a large Dataset of synchronized Audio, LyrIcs and notes, automatically created using teacher-student machine learning paradigm

* Proceedings of the 19th International Society for Music Information Retrieval Conference, ISMIR, Paris, France, pp. 431-437, 2018

Access Paper or Ask Questions