NVIDIA Active Speaker Detection ST 2110 NIM

What Is NVIDIA NIM?

NVIDIA NIM™, part of NVIDIA AI Enterprise, packages AI inferencing services that integrate directly into live SMPTE ST 2110 IP broadcast pipelines. Each NIM is deployed via Helm on a Holoscan for Media cluster and provides reliable, GPU-accelerated inference within production media workflows.

NVIDIA Active Speaker Detection (ASD) ST 2110 supports the detection and identification of active speakers in live IP broadcast environments. The NIM processes uncompressed SMPTE ST 2110 video and audio, outputs per-frame speaker metadata via SMPTE ST 2110-40 ancillary data, and optionally muxes bounding-box overlays into a video output stream for preview. The NIM is deployed via Helm on a Holoscan for Media cluster with a GPU node (L40S or NVIDIA RTX PRO 6000 Blackwell Server Edition). Supported video resolutions are 720p, 1080p, and 4K UHD, audio is mono PCM L24 at 48 kHz per stream, with multiple input streams supported (one speaker per stream). During silence, audio samples should be zero, or audio thresholding can be configured via Helm. ASD identifies unique speakers and associates spoken audio to support diarization pipelines, including across complex edits and conflicting dialogue.

Each NIM consists of a container and a model and uses a CUDA-accelerated runtime for NVIDIA GPUs. NIM on Holoscan for Media is the fastest path to integrating AI into live IP broadcast pipelines at production scale.

Getting Started with NVIDIA NIM

Deploying and integrating NVIDIA NIM is straightforward thanks to our industry standard APIs. Visit the NVIDIA Active Speaker Detection ST 2110 NIM page for release documentation, deployment guides and more.

Get Help

Enterprise Support

Get access to knowledge base articles and support cases or submit a ticket.

Governing Terms

Use of the NIM container is governed by the NVIDIA Software License Agreement and the Product-Specific Terms for NVIDIA AI Products. Use of the models is governed by the NVIDIA Open Model License. Additional Information:MIT.

You are responsible for ensuring that your use of NVIDIA AI Foundation Models complies with all applicable laws.