Skip to content

PersonaTalk

Personalised visual dubbing with lip sync

Visit official site

What PersonaTalk is

The framework is described as a two-stage attention-based system for visual dubbing that keeps a speaker's style while matching lip movement to audio.

What you can do with it

  • Encode speaking style into audio features
  • Generate lip-synced facial geometry
  • Render the face with dual attention layers

Pros & cons

✓ What we like

  • English, Chinese, German, French and Japanese are listed
  • The framework is described as general rather than person-specific

! What to watch out for

  • Dubbing starts from an audio file and a face template
  • Lip sync geometry is generated from audio features

FAQ

What does it keep?

Speaker style and facial detail are preserved.

Which languages are named?

English, Chinese, German, French and Japanese.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review