Skip to navigation

For production, Holoscan for Media is designed to run on NVIDIA-Certified Systems. For full hardware, networking, and cluster platform requirements, refer to the System Requirements for Holoscan for Media.

Models

Model nameModel IDPublisher
Active Speaker Detectionactive-speaker-detectionNVIDIA

Supported Video Resolutions

ResolutionWidth × HeightNotes
4K UHD3840 × 2160Supported
1080p1920 × 1080Default
720p1280 × 720Supported

Set the resolution via video.width and video.height. The upstream sender and downstream receiver must be configured with the same resolution.

Optimized GPU Configurations

GPUPrecision
L40SFP16
NVIDIA RTX PRO 6000 Blackwell Server EditionFP16

Pod Resource Requirements

The following resources are required per Active Speaker Detection NIM media function pod:

ResourceRequirement
CPU12 cores
Memory8 GiB
Hugepages (2 Mi)8 GiB
NVIDIA GPU1

Ensure the target node has sufficient resources and that the GPU and hugepages are configured before deploying. Node selector configuration is covered in each chart’s installation page.

Software

NVIDIA Driver

PrerequisiteVersionReference
NVIDIA Graphics Driver for Linux571.21+NVIDIA Unix Drivers

Component Versions

The Active Speaker Detection NIM on Holoscan for Media uses the following software components:

ComponentVersion
CUDA12.8.1
cuDNN9.7.1.26
TensorRT10.9.0.34
Triton Inference Serverv2.56.0
DeepStream8.0
NVIDIA Media Gateway0.7.0