About this document
Dynamic Visemes for Automatic Redubbing by Romulo Guardia is a document available to read on EtoBox.
This paper presents a novel method for automatic speech redubbing using dynamic visemes, which better represent the relationship between visible articulator motion and acoustic units compared to traditional static visemes. The method constructs a phoneme graph from dynamic viseme sequences to generate word sequences that synchronize with video, addressing the limitations of static visemes in producing plausible dialogue. The research highlights the importance of visual speech modeling for applications like
- Author
- Romulo Guardia
- Language
- EN