The LST team develops its research activities in the field of automatic speech and language processing around three axes. While historically the team has worked with data-driven approaches, it is specialized in deep learning applied to language processing.

Speech recognition

Speech recognition is the process of transforming a signal into a sequence of words. From this sequence, a great deal of information can be extracted such as opinion detection, application concepts, named entities, speech analytics...

Speaker caracterization

Speaker caracterization covers several tasks around speaker voice. We are interested in speaker segmentation and clustering (splitting of audio recordings into speakers), speaker identification and verification (finding a speaker's identity), language identification and the detection of emotions.

Machine translation

Machine translation is the process of translating (switch from one language to another) a text or a sound recording with software, without human intervention.

LST - Permanent Members

Fethi BougaresFethi Bougares

Nathalie CamelinNathalie Camelin

Paul DeléglisePaul Deléglise

Nicolas DuguéNicolas Dugué

Bruno JacobBruno Jacob

Anthony LarcherAnthony Larcher

Antoine LaurentAntoine Laurent

Jérôme LehuenJérôme Lehuen

Daniel LuzzatiDaniel Luzzati

Sylvain MeignierSylvain Meignier
Professor
Simon PetitrenaudSimon Petitrenaud

Marie TahonMarie Tahon

Jane WottawaJane Wottawa

LST - Contractual Members

Adrien BardetAdrien Bardet
PhD student
Amira BarhoumiAmira Barhoumi
PhD student
Emmanuelle BillardEmmanuelle Billard
Engineer
Rémi BouvetRémi Bouvet
PhD student
Pierre-Alexandre BrouxPierre-Alexandre Broux
PhD student
Justine CarpentierJustine Carpentier

Antoine CaubrièreAntoine Caubrière
PhD student
Pierre ChampionPierre Champion
PhD student
Salima MdhaffarSalima Mdhaffar
PhD student
Manon PinelManon Pinel
PhD student
Yevhenii ProkopaloYevhenii Prokopalo
PhD student
Thomas ThebaudThomas Thebaud
PhD student
Kévin VythelingumKévin Vythelingum
PhD student

LST Projects

Current projects

SpeechBrain

Project: SpeechBrain
Date: 09/2019 - 12/2021
Funding: Autres
Call:
Partners: Mila (Canada), LIA (France), PyTorch, IBM Research AI, fluent.ai (Canada)
Author(s): Anthony Larcher, Sylvain Meignier
URL: https://speechbrain.github.io

SpeechBrain is an open-source and all-in-one speech toolkit relying on PyTorch. The goal is to create a single, flexible, and user-friendly toolkit that can be used to easily develop state-of-the-art speech technologies, including systems for speech recognition (both end-to-end and HMM-DNN), speaker recognition, speech separation, multi-microphone signal processing (e.g, beamforming), self-supervised and unsupervised learning, speech contamination / augmentation, and many others. ► Read more

GEM

Project: Gender Equality Monitor
Date: 01/2020 - 06/2023
Funding: ANR
Call: AAPG2019
Partners: INA, LIMSI, Carism, Deezer, LERASS, ENS Lyon
Author(s): Antoine Laurent, Anthony Larcher, Nathalie Camelin, Nicolas Dugué, Sylvain Meignier
URL: https://lium.univ-lemans.fr/en/gem/

Media and society have an intricate relationship. Descriptive, the media are also prescriptive and shape our perception of the world. The GEM project aims to describe the differences in representation and treatment between women and men in the media, based on the automatic analysis of large volumes of data in French contained in the INA and Deezer collections: TV, radio, print media and music collections. ► Read more

CapDiff

Project: CapDiff
Date: 06/2019 - 12/2020
Funding: Autres
Call: BPI
Partners: LAUM (France), HAAPIE (France), VOXPASS (France)
Author(s): Anthony Larcher, Justine Carpentier, Sylvain Meignier
URL: https://lium.univ-lemans.fr/en/capdiff/

The Capdif project aims at developing a differential captation system for speech signal. This system will be designed to precisely locate each speaker in a meeting room, in order to facilitate and thus improve the performance of voice technologies. ► Read more

AISSPER

Project: A​rtificial ​I​ntelligence for ​S​emantically controlled ​SPE​ech Unde​R​standing
Date: 01/2020 - 06/2023
Funding: ANR
Call: AAPG2019
Partners: LIA (France), Orkis (France)
Author(s): Nathalie Camelin, Sylvain Meignier, Nicolas Dugué, Antoine Laurent
URL: https://lium.univ-lemans.fr/en/aissper/

Effective modeling of variabilities, exhibited by human’s speech at phoneme, word, and sentence levels, is still an open research problem for SLU systems. Even more, extracting relevant keywords, themes or concept mentions from either a sentence or an entire spoken document is a difficult task even for the most advanced end-to-end systems. AISSPER aims to develop new paradigms that jointly model acoustic and semantic information for the semantic analysis of oral documents. ► Read more

EXTENSOR

Project: End-To-end Evolutive Neural Networks for Speaker Recognition
Date: 01/2020 - 12/2021
Funding: ANR
Call: AAPG2019
Partners: Eurecom (France)
Author(s): Anthony Larcher, Antoine Laurent, Marie Tahon, Sylvain Meignier
URL: https://lium.univ-lemans.fr/en/extensor/

ExTENSoR proposes fundamental research that aims to explore the potential of using end-to-end and automatically learned / evolving articial neural networks in order to overcome the limitations of hand-crafted features and network topologies that characterise the current state of the art in many fields of speech processing. ExTENSoR also aims to bring new insights to what information in speech signals is being used in order to arrive at the scores or decisions produced by the network. ExTENSoR will pursue its objectives within the context of automatic speaker recognition and anti-spoofing, two fields of speech processing research showing burgeoning interest in end-to-end, evolutive learning. ► Read more

ASSC

Project: Fine analysis of opinions in corpuses of Customer Satisfaction
Date: 10/2019 - 10/2022
Funding: Autres
Call: MMA
Partners: MMA (France)
Author(s): Nathalie Camelin, Nicolas Dugué, Sylvain Meignier, Rémi Bouvet
URL: https://lium.uni-lemans.fr/assc

The MMA group (Mutuelles du Mans Assurances), the leading player in property and casualty insurance in France, is highly involved in the analysis of customer paths. In a customer relationship strategy, MMA wants to determine the factors of satisfaction, commitment and recommendation. Its marketing department conducts satisfaction surveys to collect numerous textual feedback from customers. The volume collected is such that it is impossible to manually process all of this data. That is why the MMA group has called on LIUM's knowledge to automatically analyze opinion information. ► Read more

PolysEmY

Project: Polysemic Embeddings for Industry
Date: 01/2020 - 07/2021
Funding: RFI AltanStic 2020
Call:
Partners: SNCF (France)
Author(s): Nathalie Camelin, Nicolas Dugué
URL: https://lium.uni-lemans.fr/polysemy

The lexical resources of SNCF's technical documentation testify to the richness and specificities of the business vocabulary used within companies such as SNCF. This vocabulary is sometimes uncommon in corpuses, but according to experts, it is of major importance for the characterization of documents. ► Read more

ON-TRAC

Project: Outils Neuronaux End-to-End pour la TRAduction des Communications
Date: 01/2019 - 01/2021
Funding: ANR
Call: Generic
Partners: LIA (Avignon) (France), LIG (Grenoble) (France), Airbus (France)
Author(s): Fethi Bougares, Antoine Laurent, Anthony Larcher
URL: https://lium.univ-lemans.fr/en/on-trac/

The ON-TRAC project proposes to radically change the architectures currently used in speech translation. It is based on end-to-end neural models for machine translation and is particularly aimed at lightweight and portable speech translation applications. ► Read more

DEEP-PRIVACY

Project: DEEP-PRIVACY
Date: 01/2019 - 12/2021
Funding: ANR
Call: generic
Partners: Multispeech (France), LIA (France), Magnet (France)
Author(s): Anthony Larcher, Antoine Laurent, Marie Tahon, Pierre Champion
URL: https://lium.univ-lemans.fr/en/deep-privacy/

DEEP-PRIVACY proposes a new paradigm based on a distributed, personalized, and privacy-preserving approach for speech processing, with a focus on machine learning algorithms for speech recognition. To this end, we propose to rely on a hybrid approach: the device of each user does not share its raw speech data and runs some private computations locally, while some cross-user computations are done by communicating through a server (or a peer-to-peer network).

► Read more

SIMPÆX

Project: SIMPÆX
Date: 02/2018 - 03/2020
Funding: RFI AltanStic 2020
Call: Amorçage, Défis scientifique 2017
Author(s): Marie Tahon
URL: https://lium.univ-lemans.fr/en/simpaex/

The SIMPÆX project aims at the automatic segmentation and identification of expressive styles and speakers in a speech corpus. Indeed the extraction of features concerning the speaker, his emotional state and the social context, offers very relevant clues for various applications such as audio indexing, automatic speech recognition, speech synthesis or human-machine interactions. ► Read more

Néo

Project: Néo
Date: 02/2018 - 01/2020
Funding: RFI AltanStic 2020
Call: Amorçage, Défis scientifique 2017
Partners: Laboratoire ERIC (France), CRTT (France)
Author(s): Nicolas Dugué, Nathalie Camelin, Yannick Estève
URL: https://lium.univ-lemans.fr/en/neo/

Neo is an interdisciplinary digital humanities research project aiming at the semi-automatic detection and analysis of contemporary neology. It results from the convergence of new language practices of the modern web (creation of new words or emergence of new senses) and recent advances in natural language processing, notably via lexicographic methods. This project combines the text mining skills of computer scientists with the experience of studying the neologisms of researchers in applied linguistics. ► Read more

ALLIES

Project: Autonomous Lifelong Learning IntelligEnt Systems
Date: 12/2017 - 11/2020
Funding: EU H2020
Call: chist-era
Partners: IDIAP (Switzerland), UPC (Span), LNE (France)
Author(s): Anthony Larcher, Loïc Barrault, Fethi Bougares, Sylvain Meignier, Yevhenii Prokopalo
URL: https://projets-lium.univ-lemans.fr/allies

The goal of the ALLIES project is to encourage and demonstrate the development of autonomous systems, able to sustain performance across time according to a given learning scenario. A learning scenario defines the importance given to the performance on “past” and “present” data in the optimization (or evaluation) process. By setting the learning scenario, a human supervisor (HS) allows or forbids the system to forget. In ALLIES, an autonomous system is fully unsupervised, incrementally adapting its models but also their structure to learn or forget events according to the given learning scenario and data that it automatically collects across time. ► Read more

Antract

Project: Transdisciplinary Analysis of French Newsreels (1945-1969)
Date: 10/2017 - 09/2020
Funding: ANR
Call: Generic
Partners: INA (France), EURECOM (France), Voxolab (France), CHS (France)
Author(s): Simon Petitrenaud, Antoine Laurent, Sylvain Meignier
URL: https://lium.univ-lemans.fr/en/antract/

The general objective of the ANTRACT project is the analysis of the images and sounds produced weekly in the framework of an independent company created in 1945, les Actualités françaises (French News), over twenty five years. This major cinematographic vector, already partially worked, has never been the subject of a systematic analysis. ► Read more

PASTEL

Project: PASTEL
Date: 10/2016 - 04/2020
Funding: ANR
Call: Interactions, Robotique, Contenus / Automatique, signal 2016
Partners: Orange Lab (France), cren (France), LS2N (France)
Author(s): Vincent Bettenfeld, Nathalie Camelin, Christophe Choquet, Christophe Després, Yannick Estève, Madeth May, Salima Mdhaffar, Lahcen Oubahssi, Claudine Piau-Toffolon
URL: https://projets-lium.univ-lemans.fr/pastel

PASTEL is a research project that aims to explore the potential of real time and automatic transcriptions for the instrumentation of mixed educational situations where the modalities of the interactions can be face-to-face or online, synchronous or asynchronous. The speech recognition technology approaches a maturity level that allows new opportunities of instrumentation in pedagogical practices and their new uses. ► Read more

Past projects

C3LS

Project: Clustering and Classification on a Corpus in Specialty Language
Date: 11/2017 - 10/2019
Funding: Autres
Call: SNCF
Partners: SNCF (France)
Author(s): Nathalie Camelin, Nicolas Dugué
URL: https://lium.uni-lemans.fr/c3ls

The SNCF group is currently undergoing a digital transformation and is increasingly turning to technologies that could use Natural Language Processing (NLP). Business documentation is currently undergoing a major transformation, with businesses that are digitising themselves, becoming more mobile and with new ways of consuming information. Various internal projects have helped to initiate a transition to digital, to find the right information at the right time. Beyond the digitisation of documents, the question arises of new intelligent systems for access to content, interpretation and data entry. The aim of the project is to identify, design and evaluate solutions that can enrich current digital documentation initiatives. The fields of application are writing assistance, information retrieval and navigation in text content. ► Read more

News.bridge

Project: News.bridge
Date: 01/2018 - 06/2019
Funding: Google
Call: Digital News Innovation Funding
Partners: Deutsche Welle, Latvian News Agency, Priberam
Author(s): Sahar Ghannay, Natalia Tomashenko, Yannick Estève
URL: https://lium.univ-lemans.fr/en/news-bridge/

The NEWS-BRIDGE project will build a commercially exploitable integrated tool with a complete workflow using language technologies to facilitate and enhance multilingual news production. This tool provides for a full broadcast news translation system, rendering existing audio, video and text news content in any of the languages supported by the various external tools for transcription, translation, voice synthesis and summarization in a variety of forms and automation applications. It combines different technologies, including speech-to-text, machine translation, text-to-speech and automated summarization. It allows for customization towards the clients, offering the user the option to get the audio of online video content into their language of choice, either as a voice-over or as subtitles by a simple mouseclick. ► Read more

MAGMAT

Project: MAGMAT
Date: 10/2016 - 10/2019
Funding: DGA/DGF
Call: Rapid
Partners: Airbus D&S (France), Voxygen (France)
URL: https://lium.univ-lemans.fr/en/magmat/

Le projet MAGMAT vise à définir et mettre en œuvre une méthodologie agile et incrémentale de développement en temps contraint d’un système de traduction le la parole vers la parole. Sur le plan du développement de langues, deux objectifs principaux sont à considérer. Le premier objectif est de mutualiser drastiquement les développements de la synthèse, de la transcription et de la traduction avec une mise commun les ressources linguistiques. Le second objectif est de définir une méthodologie rendant le processus de développement de langue adapté au contexte visé. ► Read more

Blackcompass

Project: Blackcompass
Date: 04/2016 - 04/2018
Funding: Région Pays de la Loire
Call: Fonds Pays de la Loire Territoires d'Innovation
Partners: Dictanova (France), Ville de Nantes (France)
Author(s): Sylvain Meignier, Antoine Laurent
URL: https://lium.univ-lemans.fr/en/blackcompass/

The project aims to develop a software solution to valorize phone exchange content between consumers and brands by analyzing this content. ► Read more

M2CR

Project: Multilingual Multimodal Continuous Representation for Human Language Understanding
Date: 06/2015 - 06/2019
Funding: Autres
Call: Chistera
Partners: MILA (Canada), CVC (Espagne)
Author(s): Loïc Barrault, Fethi Bougares, Nathalie Camelin, Yannick Estève, Mercedes García Martínez, Sahar Ghannay, Adrien Bardet
URL: https://projets-lium.univ-lemans.fr/m2cr

Communication is one of the necessary condition to develop intelligence in living beings. Humans use several modalities to exchange information: speech, written text, both in many languages, gestures, images, and many more. There is evidence that human learning is more effective when several modalities are used. There is a large body of research to make computers process these modalities, and ultimately, understand human language. These modalities have been, however, generally addressed independently or at most in pairs. However, merging information from multiple modalities is best done at the highest levels of abstraction, which deep learning models are trained to capture. ► Read more

EUMSSI

Project: Event Understanding through Multimodal Social
Date: 11/2013 - 10/2016
Funding: EU FP7
Call: ICT-2013.4.1 Content analytics and language technologies
Partners: UPF (Spain), L3S (Germany), VSN (Spain), GFaI (Germany), IDIAP (Switzerland)
Author(s): Yannick Estève, Vincent Jousse, Sylvain Meignier, Paul Deléglise
URL: https://www.eumssi.eu/

L’objectif principal de EUMSSI est de développer des technologies d’identification et d’agrégation d’informations non structurées provenant de sources de nature très différente (vidéo, image, audio, texte) et de différentes langues (anglais, allemand, espagnol et français). Je suis responsable de l’analyse en locuteur et nous développons conjointement avec l’IDAP un système d’identification multimodale des personnes. ► Read more