Skip to content

INFP

Audio-driven head generation for two-person dialogue

Visit official site

What INFP is

INFP is described as an audio-driven interactive head generation framework for two-person dialogue that builds expressive agent video.

What you can do with it

  • Synthesise verbal and non-verbal agent video
  • Keep facial expressions and head motions in sync
  • Adapt to dialogue states without manual role switching

Application scenarios

  • A virtual agent joins a video conference
  • A teacher uses a virtual figure for a lesson
  • A service representative talks with clients

Pros & cons

✓ What we like

  • High lip-sync accuracy and rich expressions are listed
  • Multiple languages and singing are supported

! What to watch out for

  • Input needs dual-track audio and a portrait image
  • Users set up the environment and install dependencies

FAQ

What does INFP stand for?

The material lists Interactive, Natural, Flash and Person-generic.

What speed is stated?

The material states over 40 fps on an Nvidia Tesla A10.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review