Context: The primary objective of Text-to-Speech (TTS), speech conversion and speech to speech translation system is to synthesize or generate a high-quality speech signal. Typically, the quality of synthetic speech is subjectively evaluated by human listeners. This listening test aims to assess the degree of similarity to human speech rather than machine-like speech. The main […]
Description: This thesis aims to adopt data mining techniques to study commitments to ecological transition, and in particular to characterize the rise of debates on transition technologies on socionumeric networks. Social networks are an extremely interesting source of information, as the preferred medium of communication and information for protesters. Although they come with their difficulties […]
Promoting the appropriation of serious games by teachers, through a metadesign approach, based on open-source softwareStarting: 01/01/2024PhD Student: Thierry FORESTAdvisor(s): Iza MarfisiCo-advisor(s): Bertrand MarneFunding: Collectivités locales mayennaises Serious games offer important benefits for learning, but their use by teachers remains marginal. The objective of this thesis is to understand how to promote teachers’ appropriation of […]
Interpretability of pre-trained models for automatic speech processing Starting: 04/12/2023PhD Student: Félix SagetAdvisor(s): Marie TahonCo-advisor(s): Anthony Larcher & Nicolas DuguéFunding: Allocation de recherche du ministère de l’enseignement supérieur Descritpion Avec l’avènement des réseaux de neurones, les travaux en apprentissage automatique se sont éloignés des méthodes traditionnelles qui nécessitaient une grande expertise des données, couramment appelées […]
Enriched end-to-end approach with variable complexity for speech-to-textStarting: 01/10/2023PhD Student: Youness DkhissiAdvisor(s): Anthony LarcherCo-advisor(s): Stéphane Pateux, Valentin Vielzeuf, Elys Allesiardo (Orange)Funding: CIFREGlobal context and problematics of the subject Speech transcription is an essential tool in many services; it enables a user’s request formulated in natural language to be transcribed into a dialogue system, but can […]
Modèles, méthodes et outils afin de produire des Ressources Pédagogiques Virtuelles (RPV) pour l’apprentissage de gestes techniques et professionnelsStarting: 01/10/2023PhD Student: Vincent AguedaAdvisor(s): Sebastien GEORGECo-advisor(s): Ludovic HAMONFunding: CIFRELa thèse CIFRE porte sur des modèles, méthodes et outils afin de produire des Ressources Pédagogiques Virtuelles (RPV) pour l’apprentissage de gestes techniques et professionnels. La plupart des […]
Analysis and automatic evaluation of the dental surgery gesture in a virtual training environmentStarting: 01/10/2022PhD Student: Mohamed Nail HefiedAdvisor(s): Sebastien GEORGECo-advisor(s): Ludovic HAMONFunding: ANR The subject of this thesis is the design and implementation of models and tools for the automatic evaluation of dental surgery gestures according to the observation and analysis needs of practitioners. […]
Khmer Misspelling Correction Model with Deep Learning Approach to Improve Khmer Typing Accuracy in Educational ContextStarting: 01/11/2022PhD Student: Seanghort BornAdvisor(s): Madeth MayCo-advisor(s): Funding: We are moving toward digital usage in industrial 4.0, and it has become ingrained in our lives. Prior to the invention of digital devices, we kept track of information via handwriting, but […]
Resource-constrained multilingual speech translationStarting: 01/11/2022PhD Student: Hugo RiguidelAdvisor(s): Antoine Laurent (LIUM)Co-advisor(s): Anthony Larcher (LIUM), Josep Crego (Systran)Funding: CIFRE – ministère de l’Enseignement supérieur, de la Recherche et de l’InnovationContext This thesis work is in the field of automatic language processing and more specifically in automatic speech translation. The usefulness of machine translation for the community […]
Interpretability of embeddingsStarting: 01/10/2022PhD Student: Simon GuillotAdvisor(s): Antoin Laurent (LIUM) Co-advisor(s): Nicolas Dugué (LIUM), Mathieu Valette (INALCO)Funding: ANR DIGINGDescritpion Les approches de plongements lexicaux (word embedding en anglais) ont pour objectif de représenter le vocabulaire de la langue sous forme de vecteurs intelligibles par la machine. En particulier, l’une des hypothèses derrière le succès de […]
Français