LogoWTAI Navigation
Blog Post Image

Open-source digital human series, open-source lip-sync series, audio + image generation of realistic dynamic videos

Open-source digital human series, open-source lip-sync series, audio + image generation of realistic dynamic videos

Open-Source Audio-Driven Portrait Animation Projects

1. Sonic

  • Description: An open-source audio-driven portrait animation project by Tencent, suitable for long video generation. It performs well in lip synchronization, expressions, and head movements.
  • Demo: Details and demonstration

2. AniPortrait

  • Description: An open-source AI digital human tool that can generate animated-style dynamic videos based on user-uploaded photos and corresponding audio files.
  • Demo: Details and demonstration

3. JoyHallo

  • Description: An open-source Mandarin digital human project by JD, featuring smooth and natural lip expression.
  • Demo: Details and demonstration

4. TANGO

  • Description: An open-source lip-sync model specifically designed to synchronize character dialogue and gestures.
  • Demo: Details and demonstration

5. EchoMimicV2

  • Description: An open-source digital human video generation project by Alipay, generating half-body human animations compared to the V1 version.
  • Demo: Details and demonstration

6. Loopy

  • Description: Released by ByteDance, Loopy controls the expressions and actions of a character avatar through audio but is not open source.
  • Demo: Details and demonstration

7. OmniHuman-1

  • Description: An end-to-end multimodal conditional human video generation framework launched by ByteDance. It can generate human videos based on a single human image and motion signals but is not open source.
  • Demo: Details and demonstration

8. PersonaTalk

  • Description: An audio-driven visual dubbing framework by ByteDance, creating lip-sync videos with dubbing while preserving personal speaking style and facial details. Not open source.
  • Demo: Details and demonstration

9. JoyVASA

  • Description: An open-source audio-generated portrait and animal image animation project by JD Health and Zhejiang University.
  • Demo: Details and demonstration

10. FLOAT

  • Description: An audio-driven talking portrait video generation tool that enhances speech-driven emotional movements. Currently not open source.
  • Demo: Details and demonstration

11. INFP

  • Description: An audio-driven interactive head generation tool launched by ByteDance, enabling real-time voice communication between two digital humans. Not open source.
  • Demo: Details and demonstration

12. Hallo3

  • Description: The third version of Hallo, open-sourced by Baidu, generates videos of characters speaking the corresponding voice when inputting audio and character images, including lip sync, expressions, and head movements.
  • Demo: Details and demonstration

Publisher

avatar for WTAI
WTAI

2025/02/14

Categories

Newsletter

Subscribe online

Subscribe to our newsletter for the latest news and updates