NVIDIA announced that its suite of technologies, collectively known as NVIDIA ACE, is now generally available to developers. The suite is designed to bring digital humans to life using generative AI and promises advancements in areas such as gaming, customer service, healthcare, and more.
ACE now available for production deployments
According to the NVIDIA Technology Blog, ACE is packaged as an NVIDIA Neural Inference Microservice (NIM). These microservices enable high-quality natural language understanding, speech synthesis, and facial animation. Major companies including Aww Inc, Dell Technologies, Gumption, Hippocratic AI, Inventec, OurPalm, Perfect World Games, Reallusion, ServiceNow, SoulBotix, SoulShell, and Uneeq are integrating ACE into their platforms.
NVIDIA has also introduced the ACE PC NIM microservice, which can be deployed across the installed base of 100 million RTX AI PCs and laptops, available through early access.
Components and Updates
NVIDIA ACE 24.06 introduces general availability for various components within the Digital Human Technologies family, including NVIDIA Riva, NVIDIA Audio2Face, and NVIDIA Omniverse RTX. This is available through NVIDIA AI Enterprise.
The main microservices available in the NVIDIA NGC catalog and NVIDIA ACE GitHub repository are:
- Riva ASR 2.15.1: Adds a new English model with improved quality and accuracy.
- Riva TTS 2.15.1: Improves representation for multiple languages and includes a beta release of P-Flow for speech adaptation.
- Riva NMT 2.15.1: Introducing the new 1.5B Any-to-Any translation model.
- Audio2Pace 1.011: Adds blendshape customization options and improved lip sync for Metahuman characters.
- Omniverse Renderer Microservice 1.0.0: Adds new animation data protocols and endpoints.
- Animated Graph Microservice 1.0.0: Supports avatar position and facial expression animation.
- ACE Agent 4.0.0: Adds voice support for custom RAGs and prebuilt support for RAG workflows.
Early access microservices include:
- Nemotron-3 4.5B SLM 0.1.0: Designed for on-device inference while minimizing VRAM usage.
- Voice Live Portrait 0.1.0: Animate human portraits using audio.
- Voicefont 1.1.1: Reduce latency for real-time use cases and support concurrent placement across GPUs.
Developer tools and workflow
To facilitate integration and deployment of ACE technology, NVIDIA has launched new workflows and developer tools available on the NVIDIA ACE GitHub. The Kairos Gaming reference workflow includes the Audio2Face plugin for Unreal Engine 5, and the NVIDIA Tokkio Customer Service reference workflow includes a variety of tools and samples for configuring digital humans.
Additional developer tools include:
- Unified Cloud Services Tools 2.5: Simplify NVIDIA Cloud Functions application deployment.
- Avatar Configurator 1.0.0: Adds new default avatar, hairstyle, and clothing options.
ACE NIM Microservice for RTX AI PC
In addition to data center deployments, NVIDIA is delivering ACE NIM microservices to its installed base of 100 million RTX AI PCs and laptops. NVIDIA Nemotron-3 4.5B, the first small-scale language model, is designed for on-device inference with accuracy comparable to large-scale language models running in the cloud. Early access to Nemotron-3 4.5B SLM is now available, with Audio2Face and NVIDIA Riva ASR on-device models coming soon.
Developed in partnership with Inworld AI, the Covert Protocol technology demo demonstrates the capabilities of Audio2Face and Riva ASR running locally on a GeForce RTX PC.
Getting started
Developers can start working with NVIDIA ACE by evaluating the latest ACE NIMs directly in the browser or through API endpoints running on a fully accelerated stack. NVIDIA provides tools and workflows to accelerate integration and enable early access to microservices so developers can see how ACE can transform their future pipelines.
For companies seeking end-to-end digital human solutions or custom development from ACE, NVIDIA’s partner network includes service delivery partners such as Convai, Inworld AI, Data Monsters, Quantiphi, Soulshell, Top Health Tech, and UneeQ.
Developers with questions or feedback about Digital Human technologies can visit the Digital Human Forum.
Image source: Shutterstock
. . .
tag