Skip to navigation

This page lists the models, GPU architectures, and software prerequisites supported by the Active Speaker Detection NIM.

Models

The following table lists the model packaged in this NIM.

Model NameModel IDPublisher
Active Speaker Detectionactive-speaker-detectionNVIDIA

Optimized Configurations

The following table lists GPUs with optimized configurations.

GPUArchitecture (CC)Precision
NVIDIA RTX PRO 6000 Blackwell Server Edition, NVIDIA RTX PRO 4500 Blackwell Server EditionBlackwell (12.0)FP16
B200Blackwell (10.0)FP16
H100, H200Hopper (9.0)FP16
L4, L40, L40SAda (8.9)FP16
A2, A10, A16, A40Ampere (8.6)FP16
A100Ampere (8.0)FP16
T4Turing (7.5)FP16

Other Architectures (Workstation/Consumer GPUs)

The following table lists additional workstation and consumer GPUs that are supported.

GPUArchitecture (CC)Precision
NVIDIA RTX PRO 6000 Blackwell Workstation Edition, RTX 5090, RTX 5080Blackwell (12.0)FP16
RTX 6000 Ada Generation, RTX 4090Ada (8.9)FP16

The Active Speaker Detection NIM is compatible with professional and consumer GPUs that have Tensor cores and are based on the following NVIDIA architectures: Blackwell, Hopper, Ada, Ampere, and Turing. RTX-based GPUs of those architectures are also supported.

The Active Speaker Detection NIM uses NVDEC hardware acceleration for video decoding.

GPUs without NVDEC hardware support might not be supported.

Some GPUs support only a limited number of concurrent NVDEC sessions, which means the NIM can process only that same number of concurrent inputs on those GPUs.

For details, refer to the Video Encode and Decode Support Matrix.

Software

This section lists the NVIDIA driver, Docker, and software stack versions required to run Active Speaker Detection NIM.

NVIDIA Driver and Prerequisites

NVIDIA driver requirements and other prerequisites for Active Speaker Detection NIM:

PrerequisiteVersionDownload and Installation Steps
NVIDIA Graphic Drivers for Linux571.21+NVIDIA Unix Drivers
DockerlatestUbuntu, CentOS, Debian: Install Docker Engine; Rocky Linux: Docker - Install Engine
NVIDIA Container ToolkitlatestInstallation and configuration instructions

On bare-metal Linux systems with PCIe topology and multiple GPUs, we recommend setting IOMMU to passthrough. Refer to PCIe Multi-GPU Systems.

Active Speaker Detection NIM uses the following NVIDIA software platforms:

ComponentsVersion
CUDA12.8.1
cuDNN9.7.1.26
TRT10.9.0.34
Triton Inference Serverv2.56.0
DeepStream8.0