Post-doctoral Research Visit F/m Deep Generative

Il y a 2 mois


Strasbourg, France Inria Temps plein

Le descriptif de l’offre ci-dessous est en Anglais_

**Type de contrat **:CDD

**Niveau de diplôme exigé **:Thèse ou équivalent

**Fonction **:Post-Doctorant

**Contexte et atouts du poste**:
This postdoctoral research is part of the **REAVISE** project: “Robust and Efficient Deep Learning based Audiovisual Speech Enhancement” (2023-2026) funded by the French National Research Agency (ANR). The general objective of REAVISE is to develop a unified audio-visual speech enhancement (AVSE) framework. This will leverage recent breakthroughs in statistical signal processing, machine learning, and deep neural networks to create a robust and efficient AVSE system.
- The postdoctoral researcher will be supervised by Mostafa Sadeghi (researcher, Inria), Romain Serizel (associate professor, University of Lorraine), as members of the Multispeech team, and Xavier Alameda-Pineda (Inria Grenoble), member of the RobotLearn team. The team has access to powerful computational resources, including efficient GPUs and CPUs, required for the experiments planned in this project.**Work environment**:Multispeech team, Inria Nancy, France.**Starting date & duration**:October 2024 (flexible), for two years.

**Mission confiée**:
**Background.** Audio-visual speech enhancement (AVSE) aims to improve the intelligibility and quality of noisy speech signals by utilizing complementary visual information, such as the lip movements of the speaker [1]. This technique is especially useful in highly noisy environments. The advent of deep neural network (DNN) architectures has led to significant advancements in AVSE, prompting extensive research into the area [1]. Existing DNN-based AVSE methods are divided into supervised and unsupervised approaches. In supervised approaches, a DNN is trained on a large audiovisual corpus, like AVSpeech [2], which includes a wide range of noise conditions. This training enables the DNN to transform noisy speech signals and corresponding video frames into a clean speech estimate. These models are typically complex, containing millions of parameters.

On the other hand, unsupervised methods [3-5] employ statistical modeling combined with DNNs. These methods use deep generative models, such as variational autoencoders (VAEs) [6] and diffusion models [7], trained on clean datasets like TCD-TIMIT [8], to probabilistically estimate clean speech signals. Since these models do not train on noisy data, they are generally lighter than supervised models and may offer better generalization capabilities and robustness to visual noise, as indicated by their probabilistic nature [3-5]. Despite these advantages, unsupervised methods remain less explored compared to their supervised counterparts.

**Principales activités**:
**Objectives.** In this project, we aim to develop a robust and efficient AVSE framework by thoroughly exploring the integration of recent deep-learning architectures designed for speech enhancement, encompassing both supervised and unsupervised approaches. Our goal is to leverage the strengths of both strategies alongside cutting-edge generative modeling techniques to bridge their gap. This includes the implementation of computationally efficient multimodal (latent) diffusion models, dynamical VAEs [9], temporal convolutional networks (TCNs) [10], and attention-based methods [11]. The main objectives of the project are outlined as follows:

- Develop a neural architecture that assesses the reliability of lip images—whether they are frontal, non-frontal, occluded, in extreme poses, or missing—by providing a normalized reliability score at the output [12];
- Design deep generative models that efficiently exploit the sequential nature of data and effectively fuse audio-visual features;
- Integrate the visual reliability analysis network within the deep generative model to selectively use visual data. This will enable a flexible and robust framework for audio-visual fusion and enhancement.

**References**:

- [1] D. Michelsanti, Z. H. Tan, S. X. Zhang, Y. Xu, M. Yu, D. Yu, and J. Jensen, "An overview of deep learning-based audio-visual speech enhancement and separation," in _IEEE/ACM Transactions on Audio, Speech, and Language Processing_, vol. 29, 2021.
- [2] A. Ephrat, I. Mosseri, O. Lang, T. Dekel, K. Wilson, A. Hassidim, W.T. Freeman, M. Rubinstein, "Looking-to-Listen at the Cocktail Party: A Speaker-Independent Audio-Visual Model for Speech Separation," in _SIGGRAPH_ 2018.
- [3] M. Sadeghi, S. Leglaive, X. Alameda-Pineda, L. Girin, and R. Horaud, "Audio-visual speech enhancement using conditional variational auto-encoders," in _IEEE/ACM Transactions on Audio, Speech and Language Processing_, vol. 28, pp. 1788 -1800, 2020.
- [4] A. Golmakani, M. Sadeghi, and R. Serizel, "Audio-visual Speech Enhancement with a Deep Kalman Filter Generative Model," in _IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP)_, Rhodes Island, June 2023.
- [5] B. Nortier, M. Sadeghi, and



  • Strasbourg, France Université de Bourgogne-Franche-Comté Temps plein

    **Intelligence Artificielle générative pour la médiation culturelle et patrimoniale des archives numériques**: - Réf **ABG-123672** - Sujet de Thèse- 27/05/2024- Contrat doctoral- Université de Bourgogne-Franche-Comté- Lieu de travail- Montbéliard - Bourgogne-Franche-Comté - France- Intitulé du sujet- Intelligence Artificielle générative pour...


  • Strasbourg, France Inria Temps plein

    Le descriptif de l’offre ci-dessous est en Anglais_ **Type de contrat **:CDD **Niveau de diplôme exigé **:Thèse ou équivalent **Fonction **:Post-Doctorant **Niveau d'expérience souhaité **:Jeune diplômé **Contexte et atouts du poste**: This postdoc would be within the MFX team. The topic is associated to a competitve call, where a jury...


  • Strasbourg, France Inria Temps plein

    Le descriptif de l’offre ci-dessous est en Anglais_ **Type de contrat **:CDD **Niveau de diplôme exigé **:Thèse ou équivalent **Fonction **:Post-Doctorant **Contexte et atouts du poste**: [1] “Fragment-based modelling of single stranded RNA bound to RNA recognition motif containing proteins” Isaure Chauvot de Beauchêne, Sjoerd De Vries, Martin...


  • Strasbourg, France Inria Temps plein

    Le descriptif de l’offre ci-dessous est en Anglais_ **Type de contrat **:CDD **Niveau de diplôme exigé **:Thèse ou équivalent **Fonction **:Post-Doctorant **Mission confiée**: **Principales activités**: **Avantages**: - Subsidized meals - Partial reimbursement of public transport costs - Leave: 7 weeks of annual leave + 10 extra days off due to...


  • Strasbourg, France IGBMC Temps plein

    The team Transcriptional regulation of neural and immune development at the IGBMC (Institut de génétique et de biologie moléculaire et cellulaire) in Strasbourg is recruiting a postdoctoral researcher to study the cross-talk between the macrophages and the intestine epithelium during enteric infection using the model Drosophila melanogaster. The project...


  • Strasbourg, France IGBMC Temps plein

    The team Transcriptional regulation of neural and immune development at IGBMC (Institut de Génétique et de Biologie Moléculaire et Cellulaire) in Strasbourg is recruiting a postdoctoral researcher to study macrophage diversity using the model Drosophila melanogaster. The project is funded by an ANR grant and is carried out in close collaboration with the...


  • Strasbourg, France CNRS Temps plein

    Cette offre est disponible dans les langues suivantes: - Français - Anglais Date Limite Candidature : samedi 5 août 2023 **Informations générales**: **Intitulé de l'offre **:Chercheur H/F post-doctoral en thermochronologie (double datation U-Th/He et U-Pb)** Référence : UMR7358-JULCHA-001 Nombre de Postes : 1 Lieu de travail : VANDOEUVRE LES...


  • Strasbourg, France CNRS Temps plein

    Cette offre est disponible dans les langues suivantes: - Français - Anglais Date Limite Candidature : jeudi 15 juin 2023 **Informations générales**: **Intitulé de l'offre **:Contrat doctoral en automatique / théorie du contrôle (H/F) - Contrôle des systèmes interconnectés par programmation dynamique : stabilité et robustesse** Référence :...

  • Post-doc (H/F)

    il y a 1 mois


    Strasbourg, France CNRS Temps plein

    Cette offre est disponible dans les langues suivantes: - Français - Anglais Date Limite Candidature : mardi 5 septembre 2023 **Informations générales**: **Intitulé de l'offre **:Post-doc (H/F) _ Erosion verticale et latérale des rivières à fond rocheux : quantification des processus à partir de mesures expériementales et en rivières naturelles...


  • Strasbourg, France ArcelorMittal Temps plein

    ArcelorMittal is the world’s largest steel producer. We use the most innovative technology to create the steels tomorrow’s world will be made of. Every day over 190,000 of our talented people, located in over 60 countries, push the boundaries of digitalization and use advanced technology to create a world that is stronger, faster and smarter. To help...


  • Strasbourg, Grand Est, France CNRS Temps plein

    10 Dec 2023Job Information Organisation/CompanyCNRSDepartmentInstitut de Chimie et Procédés pour l'Energie, l'Environnement et la SantéResearch FieldChemistryPhysicsTechnologyResearcher ProfileFirst Stage Researcher (R1)CountryFranceApplication Deadline30 Dec 2023 – 23:59 (UTC)Type of ContractTemporaryJob StatusFull-timeHours Per Week35Offer Starting...


  • Strasbourg, France ArcelorMittal Temps plein

    ArcelorMittal is the world’s largest steel producer. We use the most innovative technology to create the steels tomorrow’s world will be made of. Every day over 190,000 of our talented people, located in over 60 countries, push the boundaries of digitalization and use advanced technology to create a world that is stronger, faster and smarter. To help...


  • Strasbourg, France Universitat Politècnica de València Temps plein

    **Prediction of biosourced concrete properties through artificial intelligence models**: - Réf **ABG-119898** - Sujet de Thèse- 01/02/2024- Autre financement privé- Universitat Politècnica de València- Lieu de travail- Strasbourg - Grand Est - France- Intitulé du sujet- Prediction of biosourced concrete properties through artificial intelligence...


  • Strasbourg, France CNRS Institut Charles Sadron Temps plein

    **SYNTHESE ORGANIQUE DE SONDES CHIRALES MECANO-SENSIBLES**: - Réf - **ABG-117905** - Stage master 2 / Ingénieur- Durée 4 mois- Salaire net mensuel 80000- 15/11/2023- CNRS Institut Charles Sadron- Lieu de travail- Strasbourg Grand Est France- Champs scientifiques- Chimie - Mots clés- Organic Chemistry, Atropisomer, Soft Mechanochemistry, Chiroptical...


  • Strasbourg, France Inria Temps plein

    Le descriptif de l’offre ci-dessous est en Anglais_ **Type de contrat **:CDD**Niveau de diplôme exigé **:Bac + 5 ou équivalent**Autre diplôme apprécié **:PhD**Fonction **:Ingénieur scientifique contractuel**Niveau d'expérience souhaité **:De 5 à 12 ansContexte et atouts du poste **Team** Mission confiée **Context** In the context of a...


  • Strasbourg, France IGBMC Temps plein

    New postdoc and doctoral positions available in the laboratory of Romeo Ricci at the IGBMC in Strasbourg, France (; ). Our institute is a leading research institution in Europe currently hosting 46 teams with researchers from all over the world. It is a facility-driven institute with state-of-the-art equipment and core expertise covering all experimental...


  • Strasbourg, France AGF DU BAS RHIN Temps plein

    L’Association Générale des Familles (AGF) du Bas-Rhin recherche un(e) **Chargé-e de Visites Médiatisées - Coordinateur-rice des Visites Médiatisées H/F **en Contrat à Durée Déterminé (CDD) à temps plein avec possible évolution vers un Contrat à Durée Indéterminée (CDI). **Temps de travail**: - Semaine 1 : du mardi au samedi -> 3 jours au...


  • Strasbourg, France NovAliX Temps plein

    Starting date: ASAP Type of contract: Temporary Location: Strasbourg, FRANCE Job offer: Project manager junior in organic chemistry Job description You will be working in the chemistry department under the responsibility of a team leader or a project manager senior. Your work will consist in: - Conducting chemistry programs for the delivery of new...

  • Visiteur Médical

    il y a 1 mois


    Strasbourg, France SOFIP Temps plein

    Prêt(e) pour un nouveau challenge ? Vous aimez travailler dans un environnement innovant et recherchez une entreprise qui vous permet d’évoluer rapidement ? Depuis 40 ans, SOFIP accompagne les marques santé / bien-être / cosmétique dans leur développement commercial en officine, parapharmacie et visite médicale. Les plus grands laboratoires nous...


  • Strasbourg, France NovAliX Temps plein

    Starting date: Asap Type of contract: Permanent Location: Illkirch (67, near Strasbourg) Job offer: Research Assistant Compounds Management DEL M/F Job description You will be mainly working with the DEL chemistry team, dedicated to the synthesis of DNA encoded libraires and involved in Drug Discovery projects. Your work will consist in: - Managing and...