Hi everyone,
I’m planning to use a Convai AI character inside a vertical Holobox display (1080x1920 resolution), the one that i have used in Event was Convai Call and I’m aiming for the highest possible visual fidelity and smooth interaction.
Among the available options — Unreal Engine, Unity, NVIDIA Omniverse, or Convai Call — which platform provides the best video output quality and control over the 3D character rendering?
Has anyone tested these in a similar setup? Any tips on resolution handling, lip-sync clarity, or lighting control would be greatly appreciated.
If your priority is high visual fidelity, smooth interaction, and minimizing development time, here’s a comparison of your options:
Convai Avatar Studio
Best suited for ease of setup and quality.
Supports custom character import (including MetaHumans or Reallusion avatars).
Lipsync is client-side, making it more precise and visually aligned.
The experience runs over Pixel Streaming, meaning it doesn’t require powerful local hardware. Visual quality will depend on your internet connection, not client specs.
Unreal Engine
Offers the most control over lighting, shaders, and overall rendering fidelity.
However, lip-sync is server-side and delivered with the response, which may result in less accurate sync.
Requires you to handle all rendering and animation logic manually.
Ideal if you need fine-grained control and have access to high-end hardware.
Recommendation
Use Convai Avatar Studio if you want a quick setup with good lipsync and minimal hardware needs.
Choose Unreal Engine if you need maximum visual control and have the technical resources to fine-tune everything.
Thank you for quick reply. I’ve used Convai Avatar Studio into our Holobox (vertical 1080x1920) display, and it looks great visually.
However, there’s one challenge preventing us from offering it fully to end users:
Currently, voice interaction requires pressing T on the keyboard to activate the microphone. Since Holobox is designed as a standalone user-facing unit, we need to remove the keyboard and make this more seamless.
Question:
Is there any way to:
Auto-activate the microphone when the user is speaking?
Replace the T keypress with a touchscreen button or external trigger (e.g., motion sensor, button panel)?
Any webhook or JS API options to trigger the microphone programmatically?
We’re looking for a keyboard-free, intuitive solution for real-world deployment. I’d really appreciate guidance from the team or anyone who has solved something similar.
Thank you for the update!
That’s great to hear — hands-free support will be a game changer, especially for interactive kiosk setups like Holobox. Looking forward to the integration, and please let me know if there’s any way I can help with early testing or feedback.
Sounds great — thank you!
I’ll keep an eye out for the update. This feature will make a big difference in creating a more seamless and user-friendly experience with Holobox. Appreciate your efforts!