DVPS

AI & ML interests

Multimodal foundation models. The science, technology and applications. Domains: language, medicine, environment.

Recent Activity

spapi  authored a paper 7 days ago

FAMA: The First Large-Scale Open-Science Speech Foundation Model for English and Italian

spapi  authored a paper 7 days ago

Better Late Than Never: Evaluation of Latency Metrics for Simultaneous Speech-to-Text Translation

mgaido91  authored a paper 8 days ago

Cross-Attention is Half Explanation in Speech-to-Text Models

View all activity

Team members 10