MetaHuman xADA for Custom Audio Streams [ Guidance Needed ]

Development Character & Animation
1

I’m building a production system using MetaHumans with real-time AI-generated voice , and I need architectural guidance on the best path forward with xADA (Audio Driven Animation).

Current Situation:

  • Using UE 5.7 with pixel streaming architecture

  • Need to drive MetaHuman lip sync from streamed audio (not physical microphones)

  • Audio source: ElevenLabs API delivering PCM audio in chunks

I’ve researched the MetaHuman Animator documentation extensively, and I understand xADA offers two paths:

  1. Real-time via MetaHuman Audio Live Link Source - Designed for physical USB audio devices

  2. Offline via MetaHuman Performance assets - Requires complete SoundWave assets upfront

Neither path seems architected for streaming audio from external APIs, which is increasingly common for AI-powered characters.

What I’ve Discovered:

  • MetaHumanLocalLiveLinkSource API is not exposed for extension

  • Live Link presets store device UUIDs, not flexible audio routing

  • The offline path works brilliantly but requires the complete audio file first

  • Third-party plugins exist, but we specifically need Epic’s native xADA quality and long-term support

My Questions:

  1. Is there a supported approach to create a custom Live Link Audio Source that feeds into xADA’s processing pipeline? If so, which classes/interfaces should I extend?

  2. Could you clarify the architecture of how audio flows from MetaHuman Audio Live Link Source → xADA processing → facial animation? Understanding this would help me determine if a custom source is feasible.

  3. For runtime SoundWave generation, what’s the recommended pattern for creating SoundWave assets from PCM data that xADA’s offline processor will accept? Should I use USoundWaveProcedural, or is there a better approach?

  4. Is there any roadmap consideration for supporting programmatic audio input to xADA? The use case of AI-driven charactersis becoming extremely common, and native support would be incredibly valuable.

Thank you for you time!