Picture for Romain Hennequin

Romain Hennequin

Singing Language Identification using a Deep Phonotactic Approach

May 31, 2021
Lenny Renault, Andrea Vaglio, Romain Hennequin

Figure 1 for Singing Language Identification using a Deep Phonotactic Approach

Figure 2 for Singing Language Identification using a Deep Phonotactic Approach

Figure 3 for Singing Language Identification using a Deep Phonotactic Approach

Figure 4 for Singing Language Identification using a Deep Phonotactic Approach

* ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 271-275
* 5 pages, 1 figure, ICASSP 2021

Access Paper or Ask Questions

Modeling the Music Genre Perception across Language-Bound Cultures

Oct 13, 2020
Elena V. Epure, Guillaume Salha, Manuel Moussallam, Romain Hennequin

Figure 1 for Modeling the Music Genre Perception across Language-Bound Cultures

Figure 2 for Modeling the Music Genre Perception across Language-Bound Cultures

Figure 3 for Modeling the Music Genre Perception across Language-Bound Cultures

Figure 4 for Modeling the Music Genre Perception across Language-Bound Cultures

* 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP 2020)

Access Paper or Ask Questions

Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation

Sep 16, 2020
Elena V. Epure, Guillaume Salha, Romain Hennequin

Figure 1 for Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation

Figure 2 for Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation

Figure 3 for Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation

Figure 4 for Multilingual Music Genre Embeddings for Effective Cross-Lingual Music Item Annotation

* 21st International Society for Music Information Retrieval Conference (ISMIR 2020)

Access Paper or Ask Questions

Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

Aug 26, 2020
Darius Afchar, Romain Hennequin

Figure 1 for Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

Figure 2 for Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

Figure 3 for Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

Figure 4 for Making Neural Networks Interpretable with Attribution: Application to Implicit Signals Prediction

* 14th ACM Conference on Recommender Systems (RecSys '20)

Access Paper or Ask Questions

Simple and Effective Graph Autoencoders with One-Hop Linear Models

Jan 21, 2020
Guillaume Salha, Romain Hennequin, Michalis Vazirgiannis

Figure 1 for Simple and Effective Graph Autoencoders with One-Hop Linear Models

Figure 2 for Simple and Effective Graph Autoencoders with One-Hop Linear Models

Figure 3 for Simple and Effective Graph Autoencoders with One-Hop Linear Models

Figure 4 for Simple and Effective Graph Autoencoders with One-Hop Linear Models

* Under review. A preliminary version of this work has previously been presented at a NeurIPS 2019 workshop without proceedings: arXiv:1910.00942

Access Paper or Ask Questions

Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

Oct 02, 2019
Guillaume Salha, Romain Hennequin, Michalis Vazirgiannis

Figure 1 for Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

Figure 2 for Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

Figure 3 for Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

Figure 4 for Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

* NeurIPS 2019 Graph Representation Learning Workshop

Access Paper or Ask Questions

Improving Collaborative Metric Learning with Efficient Negative Sampling

Sep 24, 2019
Viet-Anh Tran, Romain Hennequin, Jimena Royo-Letelier, Manuel Moussallam

Figure 1 for Improving Collaborative Metric Learning with Efficient Negative Sampling

Figure 2 for Improving Collaborative Metric Learning with Efficient Negative Sampling

* SIGIR 2019

Access Paper or Ask Questions