George Sterpu

PhD Student

Status: Former Member

Email Address: [email protected]

George is a PhD student working on the problem of multimodal fusion for Audio-Visual Speech Recognition (AVSR). George’s main research interests are Multimodal Interaction, Speech Processing, and Machine Learning. Some of George’s significant original contributions to knowledge include a multimodal integration algorithm for coherent but asynchronous modalities (AV Align), and a real-time speech decoding algorithm that aims to segment words in fluent speech (Taris).

Research Domains
  • Artificial Intelligence
  • Audio
  • Deep Learning
  • Image and Video processing
  • Machine Learning
  • Multimodal Interaction
  • Speech