Pierre-Alexandre Broux

PhD defence, Pierre-Alexandre Broux Date : 10/01/2020 Time : 14h00 Location : Room 210, IC2 building, LIUM, Le Mans Université Title : Speaker diarization in audiovisual files in interaction with human annotators Jury members : Reviewers: – Jean-François BONASTRE (LIA, Université d’Avignon) – Nicholas EVANS (EURECOM) Examiners: – Régine ANDRE-OBRECHT (Université Toulouse 3) Supervisor: – […]

LST half-day

LST half-day   Le 20 décembre aura lieu la demi-journée de l’équipe LST. A cette occasion, les chercheurs —jeunes et plus confirmés — présenteront leurs thématiques de recherche.   Après la présentation de l’équipe par son responsable, Sylvain Meignier, les deux axes de recherche, locuteur et langage, seront introduits par Anthony Larcher et Natalie Camelin […]

Kevin Vythelingum

PhD defence, Kevin Vythelingum Date : 10/12/2019 Heure : 13h30 Lieu : Board room, IC2 building, LIUM, Le Mans Université Title : Rapid, efficient and joint construction of speech recognition and synthesis systems for new languages Jury members : Reviewers: – Martine ADDA-DECKER (LPP, Université Paris 3 Sorbonne) – Denis JOUVET (LORIA, INRIA Nancy) Examiners: […]

Anthony Larcher, speaker invité au workshop Life Long Learning d’ASRU 2019

Anthony Larcher is an invited speaker at the Life Long Learning for Spoken Language Systems Workshop colocated with ASRU 2019 in Singapore from 14 to 18th of December. The workshop will bring together experts in spoken language systems whose research focuses on solving problems related to continual improvement of speech processing systems such as conversational […]

Participation du LIUM à Campus Sonore

LIUM’s participation in Campus sonore   During the international sound biennial “Le Mans sonore” LIUM participates in Campus sonore with demonstrations of speech, translation and speaker recognition on 7 and 8 December. The speech signal contains a lot of information, some of which is related to the identity of the speaker. For this day, LIUM […]

Plongements lexicaux : modèles, temporalité, biais, interprétabilité

Seminars from Nicolas Dugué, lecturer at LIUM   Date: 22/11/2019 Hour: 11h00 Localisation: IC2, Room 210 Speaker: Nicolas Dugué     Word embeddings : models, time-based generalization, gender bias, interpretability After an introduction to word embeddings models, we shall discuss the challenges related to their use in practice. First, we shall describe diachronic time-based word […]

Participation du LIUM à Voice Tech Paris 2019 avec AlloMédia

LIUM’s participation in Voice Tech Paris 2019 with AlloMedia   LIUM is participating, with AlloMedia, in Voice Tech Paris 2019, the 1st B2B event dedicated to voice technologies in France, which will be held on 26 and 27 November 2019. Workshop: What marketing doesn’t tell you about automatic speech recognition What are the limitations of […]

A Study on Multilingual Transfer Learning in Neural Machine Translation : Finding the Balance Between Languages / An Empirical Evaluation of Arabic-specific Embeddings for Sentiment Analysis.

Seminars from Adrien Bardet and Amira Barhoumi, PhD students at LIUM   Date: 08/11/2019 Heure: 11h00 Lieu: IC2, Salle des conseils Intervenant: Adrien Bardet, Amira Barhoumi   A Study on Multilingual Transfer Learning in Neural Machine Translation : Finding the Balance Between Languages Transfer learning is a solution to the problem of lack of data […]

Campagne d’évaluation IWSLT: 1ere place dans la tâche de traduction de la parole

IWSLT – results – First place End-to-End speech translation   The LIUM participated to the evaluation campaign organized by the 16th International Workshop on Spoken Language Translation (IWSLT). ON-TRAC Consortium (LIG, LIA, LIUM) participated in the Speech Translation task, (English-Portuguese sub-task) and its end-to-end system ranked first. More details: https://zenodo.org/record/3525578

Evaluation of lifelong learning systems

For continuous use intelligent systems needs the support of machine learning experts, what is expensive. To reduce costs it is reasonable to implement long life learning intelligent systems. Idea of this type of systems consists in continuous adaptation of system for new conditions of execution. This allows to stay independent from machine learning expert intervention, […]