HDR, Antoine Laurent

HDR Defence, Antoine Laurent Date : 14/01/2022 Time : 14h00 Localization : Online   Title : Resource-constrained speech recognition and its applications   Jury members: Denis Jouvet (PR, LORIA, Nancy), Reviewer Georges Linarès (PR, LIA, Avignon), Reviewer Benoit Favre (PR, LIF, Marseille), Reviewer Laurent Besacier (PR, NLE, Grenoble) Richard Dufour (PR, LS2N, Nantes) Anthony Larcher […]

Privacy-Preserving Speech Representation Learning using Vector Quantization

Seminar from Pierre Champion, Phd student at INRIA of Nancy and LIUM   Date: 14/01/2022 Time: 11h00 Localization: IC2 Boardroom, online Speaker: Pierre Champion   Privacy-Preserving Speech Representation Learning using Vector Quantization   Speech signals are a rich source of speaker-related information, including sensitive attributes like gender, identity, etc. Those sensitive attributes can be extracted […]

Offre de stage M2 – Deep learning pour l’holographie numérique

Deep learning pour l’holographie numérique Supervisors: Marie Tahon, Silvio Montrésor Hosting lab : LIUM – LAUM Place : Le Mans Contact : Marie.Tahon(at)univ-lemans.fr, Silvio.Montresor(at)univ-lemans.fr,   Contexte du stage : Au sein de l’équipe transducteur du LAUM nous travaillons actuellement en collaboration avec le LIUM sur le développement d’un réseau de neurones profond en python pour […]

DIGING

Dynamic and Interpretable Graph-based word embeddINGs (DIGING)Date: 01/2022 – 12/2025Funding: ANRCall: ANR JCJCURL: https://lium.univ-lemans.fr/en/diging/LIUM Participant(s): Nicolas DuguéNathalie CamelinSylvain MeignierJane WottawaSimon GuillotAnna BérangerRecent approaches to embedding learning have focused on results, often at the expense of interpretability and algorithmic complexity. Yet, interpretability is a necessary pre-requisite for the deployment of such technologies when they are used […]

Summarising Historical Text in Modern Languages

Seminar from Chenghua Lin, lecturer at the University of Sheffield   Date: 07/01/2022 Time: 11h00 Localization: online Speaker: Chenghua Lin   Summarising Historical Text in Modern Languages   We introduce the task of historical text summarisation, where documents in historical forms of a language are summarised in the corresponding modern language. This is a fundamentally […]

Journée internationale de sensibilisation des jeunes filles aux métiers scientifiques et techniques

International day for Girls’ Science and Technology Awareness   On Thursday 9 December, the international day to raise awareness among young girls about scientific and technical professions took place. Nathalie Camelin, Nicolas Dugué and Marie Tahon spoke to the students of the Touchard high school in partnership with the association “Elles bougent”. During this afternoon, […]

Extending the study of speech accommodation: the relationship between hypospeech and speakers’ social proximity.

Seminar by Mélanie Lancien, from the university of Lausanne   Date: 03/12/2021 Time: 11h00 Localization: IC2, Boardroom, online Speaker: Mélanie Lancien   Extending the study of speech accommodation: the relationship between hypospeech and speakers’ social proximity   In linguistics, work on speech variation, mainly that in line with the H&H (Hyper-Hypoarticulation Theory) and CAT (Communication […]

carrières en recherche

Seminar from Sylvain Meignier   Date: 26/11/2021 Time: 14h00 Localization: IC2, Boardroom, online Speaker: Sylvain Meignier   Research Careers   We will discuss the follow-up of a thesis, scientific publication, preparation for qualifications and competitions in higher education and research (University, CNRS, INRIA, etc.)   

Non-Parametric Subspace Models for Acoustic Units Discovery

Seminar from Lucas Ondel, post-doc at l’Université Paris-Saclay   Date: 12/11/2021 Time: 11h30 Localization: IC2, Boardroom, online Speaker: Lucas Ondel   Non-Parametric Subspace Models for Acoustic Units Discovery   This talk will be about subspace non-parametric models for the task of learning a set of acoustic units from unlabeled speech recordings. I will show how […]