# Maxine NIM Overview NVIDIA Maxine NIMs are a collection of GPU-accelerated inference microservices that bring AI-powered video and audio effects behind consistent service interfaces. Each NIM packages a model as a self-hosted, easily deployable container, so you can add effects such as eye contact, relighting, lip sync, speech enhancement and noise removal to live and recorded media workflows. Holoscan for Media variants deploy the same capabilities into live production environments over SMPTE ST 2110. ## Use Cases Maxine NIMs span video enhancement, video analysis, speech-driven animation, audio enhancement and live production. ### Video Enhancement Use these NIMs when you need to improve or restyle video: Redirects a user's eye position toward the camera to simulate natural eye contact. Applies lighting effects to videos using HDR environment maps. Upscales video to higher resolutions for ST 2110 and compressed-video workflows. **Benefits**: Self-hosted GPU-accelerated containers, gRPC service interfaces ### Video Analysis Use these NIMs when you need to understand what is in a video: Estimates the 3D pose of every tracked person in a video. Estimates the likelihood that a video has been AI-generated. Detects and identifies active speakers in a video stream from visual and diarized audio data. **Benefits**: Per-frame and per-stream analysis, self-hosted containers, gRPC service interfaces ### Speech-Driven Animation Use these NIMs when you need facial movement that follows speech: Synchronizes lip movements in videos with input speech. Creates facial animations from a portrait photo and driving audio. **Benefits**: Speech audio combined with video or a portrait photo, self-hosted containers ### Audio Enhancement Use these NIMs when you need cleaner speech audio: Enhances speech recorded through low-quality microphones to studio-recorded quality. Removes unwanted background noise from audio. **Benefits**: Self-hosted GPU-accelerated containers, gRPC service interfaces ### Holoscan for Media Use these NIMs when you need Maxine effects in live production over SMPTE ST 2110: Studio Voice speech enhancement for Holoscan for Media deployments. Background noise removal for a single PCM audio stream over SMPTE ST 2110-30. LipSync for Holoscan for Media deployments. Active speaker detection in live video streams. **Benefits**: Helm chart and Kubernetes operator deployment, SMPTE ST 2110 and NMOS integration Instructions for AI Agents: - For clean Markdown of any page, append `.md` to the page URL - For section-specific indexes, append `/llms.txt` to any section URL ## Products - [NVIDIA Eye Contact NIM](https://docs.nvidia.com/nim/maxine/eye-contact/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/eye-contact/1.5.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/eye-contact/1.5.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/eye-contact/1.5.0/basic-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/eye-contact/1.5.0/advanced-usage.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/eye-contact/1.5.0/performance-results.md) - [NVIDIA Studio Voice NIM](https://docs.nvidia.com/nim/maxine/studio-voice/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/getting-started.md) - [NIM Model Profile Table](https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/model-profile-table.md) - [Running Inference](https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/running-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/studio-voice/1.2.1/advanced-usage.md) - [NVIDIA Studio Voice on Holoscan for Media](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/1.2.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/1.2.0/getting-started.md) - [Installation](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/1.2.0/installation/index.md) - [Kubernetes Operator Chart](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/1.2.0/installation/studio-voice-h4m-operator.md) - [End-to-End Demo Chart](https://docs.nvidia.com/nim/maxine/studio-voice-h4m/1.2.0/installation/studio-voice-h4m-sample.md) - [NVIDIA Audio2Face-2D NIM](https://docs.nvidia.com/nim/maxine/audio2face-2d/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/audio2face-2d/1.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/audio2face-2d/1.0.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/audio2face-2d/1.0.0/basic-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/audio2face-2d/1.0.0/advanced-usage.md) - [Observability](https://docs.nvidia.com/nim/maxine/audio2face-2d/1.0.0/observability.md) - [NVIDIA LipSync NIM](https://docs.nvidia.com/nim/maxine/lipsync/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/lipsync/1.4.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/lipsync/1.4.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/lipsync/1.4.0/basic-inference.md) - [PCIe Multi-GPU Systems](https://docs.nvidia.com/nim/maxine/lipsync/1.4.0/advanced-usage/multi-gpu.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/lipsync/1.4.0/performance-results.md) - [NVIDIA LipSync on Holoscan for Media](https://docs.nvidia.com/nim/maxine/lipsync-h4m/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/lipsync-h4m/1.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/lipsync-h4m/1.0.0/getting-started.md) - [Common Helm Configuration](https://docs.nvidia.com/nim/maxine/lipsync-h4m/1.0.0/configuration/common.md) - [Pipeline Configuration](https://docs.nvidia.com/nim/maxine/lipsync-h4m/1.0.0/configuration/pipeline.md) - [End-to-End Demo Chart Configuration](https://docs.nvidia.com/nim/maxine/lipsync-h4m/1.0.0/configuration/demo-chart.md) - [NVIDIA Synthetic Video Detector NIM](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/2.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/2.0.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/2.0.0/basic-inference.md) - [PCIe Multi-GPU Systems](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/2.0.0/advanced-usage/multi-gpu.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/synthetic-video-detector/2.0.0/performance-results.md) - [NVIDIA Background Noise Removal NIM](https://docs.nvidia.com/nim/maxine/bnr/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/bnr/2.1.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/bnr/2.1.0/getting-started.md) - [NIM Model Profile Table](https://docs.nvidia.com/nim/maxine/bnr/2.1.0/model-profile-table.md) - [Running Inference](https://docs.nvidia.com/nim/maxine/bnr/2.1.0/running-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/bnr/2.1.0/advanced-usage.md) - [NVIDIA Background Noise Removal on Holoscan for Media](https://docs.nvidia.com/nim/maxine/bnr-h4m/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/bnr-h4m/2.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/bnr-h4m/2.0.0/getting-started.md) - [NIM Service Chart](https://docs.nvidia.com/nim/maxine/bnr-h4m/2.0.0/installation/bnr-h4m-service.md) - [Kubernetes Operator Chart](https://docs.nvidia.com/nim/maxine/bnr-h4m/2.0.0/installation/bnr-h4m-operator.md) - [End-to-End Demo Chart](https://docs.nvidia.com/nim/maxine/bnr-h4m/2.0.0/installation/bnr-h4m-sample.md) - [NVIDIA Video Super Resolution NIM](https://docs.nvidia.com/nim/maxine/vsr/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/vsr/1.0.10/overview.md) - [ST 2110 Video Workflows](https://docs.nvidia.com/nim/maxine/vsr/1.0.10/st2110/index.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/vsr/1.0.10/st2110/getting-started.md) - [Deploy with Helm](https://docs.nvidia.com/nim/maxine/vsr/1.0.10/st2110/helm.md) - [Deploy with the Operator](https://docs.nvidia.com/nim/maxine/vsr/1.0.10/st2110/operator.md) - [NVIDIA Active Speaker Detection NIM](https://docs.nvidia.com/nim/maxine/active-speaker-detection/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/active-speaker-detection/1.2.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/active-speaker-detection/1.2.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/active-speaker-detection/1.2.0/basic-inference.md) - [PCIe Multi-GPU Systems](https://docs.nvidia.com/nim/maxine/active-speaker-detection/1.2.0/advanced-usage/multi-gpu.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/active-speaker-detection/1.2.0/performance-results.md) - [NVIDIA Active Speaker Detection on Holoscan for Media](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/1.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/1.0.0/getting-started.md) - [Installation](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/1.0.0/installation/index.md) - [NIM Service Chart](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/1.0.0/installation/active-speaker-detection-h4m-service.md) - [Kubernetes Operator Chart](https://docs.nvidia.com/nim/maxine/active-speaker-detection-h4m/1.0.0/installation/active-speaker-detection-h4m-operator.md) - [NVIDIA Relighting NIM](https://docs.nvidia.com/nim/maxine/relighting/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/relighting/1.2.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/relighting/1.2.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/relighting/1.2.0/basic-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/relighting/1.2.0/advanced-usage.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/relighting/1.2.0/performance-results.md) - [NVIDIA 3D Body Pose NIM](https://docs.nvidia.com/nim/maxine/body-pose/llms.txt) - [Overview](https://docs.nvidia.com/nim/maxine/body-pose/1.0.0/overview.md) - [Getting Started](https://docs.nvidia.com/nim/maxine/body-pose/1.0.0/getting-started.md) - [Basic Inference](https://docs.nvidia.com/nim/maxine/body-pose/1.0.0/basic-inference.md) - [Advanced Usage](https://docs.nvidia.com/nim/maxine/body-pose/1.0.0/advanced-usage.md) - [Performance Results](https://docs.nvidia.com/nim/maxine/body-pose/1.0.0/performance-results.md)