News

Develop Intelligent Virtual Assistants with NVIDIA Omniverse ACE Early Access

Discuss (0)
Three avatars, including JHH.

NVIDIA just announced at CES 2023 that early access is now available for NVIDIA Omniverse Avatar Cloud Engine (ACE). Developers and teams building avatars and virtual assistants can register to join the program, which includes access to the Omniverse ACE suite of cloud-native AI microservices for faster, easier development of interactive avatars.

Early partners include Ready Player Me, whose avatar was showcased in the CES demo. With this early access program, NVIDIA is looking for developers and partners to provide feedback on the microservices, collaborate on product improvement, and help push the limits of what’s possible with lifelike, interactive, digital humans. Applicants are reviewed and approved on a rolling basis as the program continues to expand.

What is the Omniverse ACE early access program?

Methods for developing avatars often require expertise, specialized equipment, and manually intensive workflows.

To ease avatar creation, Omniverse ACE enables the seamless integration of NVIDIA AI technologies—including pre-built models, tool sets, and domain-specific reference applications. The available avatar applications are built on most engines and deployed on public or private clouds.

Video 1. CES 2023 special address showcase of Omniverse ACE early access.

The early-access program includes access to the prerelease versions of Omniverse ACE animation AI and conversational AI microservices, including:

  • 3D animation AI microservice for third-party avatars. It uses Omniverse Audio2Face generative AI to bring to life characters in Unreal Engine and other rendering tools by creating realistic facial animation from just an audio file.
  • 2D animation AI microservice, called Live Portrait, enables easy animation of 2D portraits or stylized human faces using live video feeds.
  • Text-to-speech microservice uses NVIDIA Riva TTS to synthesize natural-sounding speech from raw transcripts without any additional information, such as patterns or rhythms of speech.

Program members also get access to tooling, sample reference applications, and supporting resources, including documentation.

What use cases are ideal for the EA program?

NVIDIA Omniverse ACE, with its cloud-native microservice design and infrastructure, is best suited for applications that will be deployed from a public or private cloud, and are designed to scale.

This early access release includes containerized microservices, which require developers to deploy through their own cloud infrastructure. Additional resources include NVIDIA UCF Studio, a low-code compute framework for developing cloud-native, real-time, and multi-modal AI applications. Sign up to apply for UCF early access.

For example, you can create a simple workflow leveraging UCF Studio to connect the Riva TTS microservice with Omniverse ACE 3D animation AI microservice. The result is a cloud-ready avatar application that speaks the inputted text with a synthetic voice and animates the avatar’s lips and facial movements.

That said, not all applications must be run on the cloud. Anyone authoring offline avatars that don’t require scaled deployment can leverage NVIDIA AI SDKs and tools, including Omniverse Audio2Face.

Get started with Omniverse ACE early access

Interested in Omniverse ACE but not sure if the early access program is the right fit? Check out the following resources to learn more and start trying out the core NVIDIA AI technologies