MetaHuman UE5.5 Preview Release

General Announcements
1

Audio Driven Animation

MetaHuman Animator can now create realistic facial animations from recorded voice audio. The process supports various languages, voices, and expressions, all within the existing, familiar MHA workflow.

The offline process is quick and easy and targets the full facial description standard. It delivers accurate and emotive lip synced animation as well as inferring upper face movement, delivering animation on the full MetaHuman facial description standard. This is a fully local solution that needs only the performance audio to bring MetaHuman characters to life.

Optimized MetaHumans in UE

We’ve applied the optimization techniques used to bring MetaHumans to UEFN, and are making them available to UE users as well. Just like in UEFN, UE users can request High, Medium and Low optimized MetaHumans and receive significant size reductions and performance improvements. This update also introduces a new ‘MetaHuman Component’ that simplifies and consolidates optimization options for MetaHumans once in UE.

Known Issues

  • Thumbnails are not updated in MetaHuman Creator
  • Sometimes a disconnect occurs when navigating back to “My MetaHumans” in MetaHuman Creator
1 Like
2

Ignore this post - I think I found the answer

3

Awesome! Will the feature support realtime generation? I’m prototying an MetaHuman AI assistant where the dialogue clips are dynamically generated via OpenAI TTS.

4

in the description of the roadmap it is written
" * A fully local, offline solution, suitable for batch-processing audio."

Has anyone tried to implement a queue for a set of audio files?