no code implementations • NAACL (NLP4IF) 2021 • Tulika Bose, Irina Illina, Dominique Fohr
Rapidly changing social media content calls for robust and generalisable abuse detection models.
no code implementations • EMNLP (insights) 2020 • Ashwin Geet D’Sa, Irina Illina, Dominique Fohr, Dietrich Klakow, Dana Ruiter
In this paper, label propagation-based semi-supervised learning is explored for the task of hate speech classification.
no code implementations • LREC 2022 • Imran Sheikh, Emmanuel Vincent, Irina Illina
Training of LSTM LMs in such limited data scenarios can benefit from alternate uncertain ASR hypotheses, as observed in our recent work.
no code implementations • LREC 2022 • Nicolas Zampieri, Carlos Ramisch, Irina Illina, Dominique Fohr
In this article, we present joint experiments on these two related tasks on English Twitter data: first we focus on the MWE identification task, and then we observe the influence of MWE-based features on the HSD task.
no code implementations • NAACL (SocialNLP) 2021 • Tulika Bose, Irina Illina, Dominique Fohr
The state-of-the-art abusive language detection models report great in-corpus performance, but underperform when evaluated on abusive comments that differ from the training scenario.
no code implementations • JEP/TALN/RECITAL 2022 • Nicolas Zampieri, Carlos Ramisch, Irina Illina, Dominique Fohr
L’identification des expressions polylexicales (EP) dans les tweets est une tâche difficile en raison de la nature linguistique complexe des EP combinée à l’utilisation d’un langage non standard.
no code implementations • 31 Jul 2023 • Nicolas Furnon, Romain Serizel, Slim Essid, Irina Illina
Speech enhancement in ad-hoc microphone arrays is often hindered by the asynchronization of the devices composing the microphone array.
no code implementations • 17 Oct 2022 • Tulika Bose, Irina Illina, Dominique Fohr
The concerning rise of hateful content on online platforms has increased the attention towards automatic hate speech detection, commonly formulated as a supervised classification task.
no code implementations • COLING 2022 • Tulika Bose, Nikolaos Aletras, Irina Illina, Dominique Fohr
State-of-the-art approaches for hate-speech detection usually exhibit poor performance in out-of-domain settings.
1 code implementation • LREC 2022 • Dana Ruiter, Liane Reiners, Ashwin Geet D'Sa, Thomas Kleinbauer, Dominique Fohr, Irina Illina, Dietrich Klakow, Christian Schemer, Angeliki Monnier
Even though hate speech (HS) online has been an important object of research in the last decade, most HS-related corpora over-simplify the phenomenon of hate by attempting to label user comments as "hate" or "neutral".
1 code implementation • Findings (ACL) 2022 • Tulika Bose, Nikolaos Aletras, Irina Illina, Dominique Fohr
In this paper, we propose to automatically identify and reduce spurious correlations using attribution methods with dynamic refinement of the list of terms that need to be regularized during training.
1 code implementation • 15 Jun 2021 • Nicolas Furnon, Romain Serizel, Slim Essid, Irina Illina
Speech enhancement promises higher efficiency in ad-hoc microphone arrays than in constrained microphone arrays thanks to the wide spatial coverage of the devices in the acoustic scene.
no code implementations • 1 Jun 2021 • Nicolas Zampieri, Irina Illina, Dominique Fohr
To incorporate MWE features, we create a three-branch deep neural network: one branch for USE, one for MWE categories, and one for MWE embeddings.
1 code implementation • 3 Nov 2020 • Nicolas Furnon, Romain Serizel, Irina Illina, Slim Essid
Deep neural network (DNN)-based speech enhancement algorithms in microphone arrays have now proven to be efficient solutions to speech understanding and speech recognition in noisy environments.
1 code implementation • 2 Nov 2020 • Nicolas Furnon, Romain Serizel, Irina Illina, Slim Essid
We propose a distributed algorithm that can process spatial information in a spatially unconstrained microphone array.
no code implementations • 2 Nov 2020 • Dominique Fohr, Irina Illina
We propose to perform this through rescoring of the ASR N-best hypotheses list.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +2
no code implementations • JEPTALNRECITAL 2020 • Rapha{\"e}l Duroselle, Denis Jouvet, Irina Illina
Sur le corpus RATS, pour sept des huit canaux radio {\'e}tudi{\'e}s, l{'}approche permet, sans utiliser de donn{\'e}es annot{\'e}es du domaine cible, de surpasser la performance d{'}un syst{\`e}me entra{\^\i}n{\'e} de fa{\c{c}}on supervis{\'e}e avec des donn{\'e}es annot{\'e}es de ce domaine.
no code implementations • JEPTALNRECITAL 2020 • Ismael Bada, Dominique Fohr, Irina Illina
Pour prendre en compte ce probl{\`e}me de prononciations erron{\'e}es, notre approche propose d{'}int{\'e}grer les prononciations non natives dans le lexique et par la suite d{'}utiliser ce lexique enrichi pour la reconnaissance.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • JEPTALNRECITAL 2020 • St{\'e}phane Level, Irina Illina, Dominique Fohr
Malgr{\'e} les avanc{\'e}s spectaculaires ces derni{\`e}res ann{\'e}es, les syst{\`e}mes de Reconnaissance Automatique de Parole (RAP) commettent encore des erreurs, surtout dans des environnements bruit{\'e}s. Pour am{\'e}liorer la RAP, nous proposons de se diriger vers une contextualisation d{'}un syst{\`e}me RAP, car les informations s{\'e}mantiques sont importantes pour la performance de la RAP.
Automatic Speech Recognition Automatic Speech Recognition (ASR) +1
no code implementations • 13 Feb 2020 • Nicolas Furnon, Romain Serizel, Irina Illina, Slim Essid
Multichannel processing is widely used for speech enhancement but several limitations appear when trying to deploy these solutions to the real-world.
no code implementations • LREC 2020 • Ashwin Geet D'Sa, Irina Illina, Dominique Fohr
The contribution of this paper is the design of binary classification and regression-based approaches aiming to predict whether a comment is toxic or not.
no code implementations • LREC 2016 • Imran Sheikh, Irina Illina, Dominique Fohr
Out-Of-Vocabulary (OOV) words missed by Large Vocabulary Continuous Speech Recognition (LVCSR) systems can be recovered with the help of topic and semantic context of the OOV words captured from a diachronic text corpus.
no code implementations • 17 Nov 2015 • Imran Sheikh, Irina Illina, Dominique Fohr, Georges Linarès
In this paper, we propose two neural network models targeted to retrieve OOV PNs relevant to an audio document: (a) Document level Continuous Bag of Words (D-CBOW), (b) Document level Continuous Bag of Weighted Words (D-CBOW2).
no code implementations • JEPTALNRECITAL 2012 • Luiza Orosanu, Denis Jouvet, Dominique Fohr, Irina Illina, Anne Bonneau