26/11/2024
Dear all,
We are excited to offer three master-level internship opportunities at the Multispeech team, Inria, Nancy, France. These projects focus on using diffusion models as data-driven speech priors and multimodal contrastive learning for speech separation, enhancement, and dereverberation. Below are the links to detailed project descriptions available online:
Diffusion-based Unsupervised Joint Speech Enhancement, Dereverberation, and Separation
Contrastive Multimodal Pretraining for Noise-aware Diffusion-based Audio-visual Speech Enhancement
Comparative Analysis of Diffusion Models and Variational Autoencoders as Data-driven Priors for Speech Enhancement
Please see the attached document for a detailed description of these projects and application procedures. Feel free to forward this email to potential candidates who may be interested.
Best regards,
Mostafa Sadeghi
Offer 1
Contrastive Multimodal Pretraining for
Noise-aware Diffusion-based Audio-visual
Speech Enhancement
General information
Supervisors Mostafa SADEGHI, Romain SERIZEL, Xavier ALAMEDA-PINEDA
Address LORIA, Campus Scientifique - BP 239, 54506 Vandœuvre-lès-Nancy
Email [email protected],[email protected]
Offer 2
Comparative analysis of diffusion models
and variational autoencoders as data-driven
priors for speech enhancement
General information
Supervisors Mostafa SADEGHI, Romain SERIZEL, Xavier ALAMEDA-PINEDA
Address LORIA, Campus Scientifique - BP 239, 54506 Vandœuvre-lès-Nancy
Email [email protected],[email protected]
Offer 3
Diffusion-based Unsupervised Joint Speech
Enhancement, Dereverberation, and
Separation
General information
Supervisors Mostafa SADEGHI, Romain SERIZEL, Xavier ALAMEDA-PINEDA
Address LORIA, Campus Scientifique - BP 239, 54506 Vandœuvre-lès-Nancy
Email [email protected],[email protected]
Send a message to learn more