Nvidia Broadcast Linux
NV Broadcast — Unofficial NVIDIA Broadcast for Linux and other OS. AI-powered virtual camera with background removal, blur, replacement, video enhancement, and noise cancellation. GPU accelerated. Open source.
Install / Use
npx skills add Hkshoonya/nvidia-broadcast-linuxInstalls into whichever agent you are using.
README
Why I Built This
I left Windows. Millions of Linux users left Windows. But we all missed one thing — NVIDIA Broadcast.
That one app that made every video call look professional. Background blur that actually worked. Virtual backgrounds that didn't look like a PowerPoint slide. Noise cancellation that silenced your mechanical keyboard.
On Linux? Nothing. You had to cobble together 5 different tools, fight with v4l2loopback configs, and still get janky edges. That's over now.
I built this because I believe Linux users deserve the same broadcast-quality experience. Not a half-baked wrapper. Not a "good enough" hack. A real, proper implementation that uses your NVIDIA GPU to do what NVIDIA Broadcast does on Windows — but open source, and in some ways, better.
"Not saying this is perfect. But I believe it will be."
This is fast. This is optimized. And the quality already rivals Windows Broadcast. With the community behind it, we'll surpass it.
— DocZeus
What's new?
See CHANGELOG.md for latest updates!
What It Does
<table> <tr> <td width="50%">Camera Effects
- Background Blur — AI-powered, person stays crystal sharp
- Background Replace — Any image via native file picker
- Green Screen — Solid green for OBS chroma key
- Auto Frame — Face tracking with Center Face and Stable Background modes
- Global Effect Hotkeys — Rebind effect toggles without bringing the app forward
- Video Enhancement — Skin smooth, enhance, sharpen, denoise, vignette
- Eye Contact Correction — Natural and Gaze Lock modes redirect gaze to camera
- Face Relighting — Fill light guided by the scene
- Recording to MP4 — NVENC hardware encode (x264 fallback)
- User Profiles — 5 built-in (Meeting, Streaming, etc.) + custom save/load
- Performance Overlay — Real-time FPS, GPU usage, VRAM, temperature
- Multi-Model — RVM (person), IS-Net (objects), BiRefNet (best edges)
- Mirror — Horizontal flip toggle
Audio Effects
- Mic Noise Removal — Kills keyboard, fan, environment noise
- Speaker Denoise — Clean up incoming audio
- Voice Effects — Bass boost, treble, warmth, compression, noise gate, gain
- 6 Voice Presets — Natural, Radio, Podcast, Deep Voice, Bright, Studio
- Microphone Selection — Full PipeWire/PulseAudio device enumeration
- Speaker Detection — All output devices via PipeWire
- Audio Level Monitor — Real-time VU meter with peak hold
- Mic Test — Record 30s / 45s / 60s and play back to test your setup
- AI Meeting Transcription — Local Whisper speech-to-text (GPU-accelerated)
- AI Meeting Summarizer — Action items, questions, key points (fully local)
- Meeting Mode — Video+audio recording with live transcription and AI summary
System Integration
- Virtual Camera — Works in Chrome, Firefox, Zoom, Discord, OBS
- Auto-Start — Launches on login, runs in background
- Setup Wizard — Auto-detects system, configures optimally
- Multi-GPU — Select which GPU runs AI effects
- Multi-Distro — Ubuntu, Fedora, Arch, openSUSE, and more
- Resolution/FPS — 360p-4K, 15-60fps, auto-detected from camera
Processing Modes
9 modes from maximum speed to CPU fallback:
| Mode | Inference | Compositing | 1080p Speed | CPU | GPU | Best For | |------|-----------|-------------|-------------|-----|-----|----------| | Killer | 360p + fused CUDA | Fused kernel (0.1ms) | 20ms / 48fps | 24% | 41% | Maximum speed | | Zeus | 480p optimized | CuPy GPU | 30ms / 33fps | 22% | 39% | Speed + quality balance | | DocZeus | 720p full quality | Fused kernel (0.1ms) | 44ms / 23fps | 22% | 46% | Best quality/speed | | CUDA Max | 720p | CuPy GPU | 45ms / 22fps | 22% | 46% | Maximum quality | | CUDA Balanced | 720p, skip 2 | CuPy GPU | 29ms / 34fps | 24% | 39% | Daily use | | CUDA Perf | 720p, skip 2 | CuPy GPU | 30ms / 34fps | 23% | 39% | Light GPU load | | CPU Quality | 720p | OpenCV SIMD | 66ms / 15fps | 17% | 27% | No CuPy fallback | | CPU Light | 720p, skip 2 | OpenCV SIMD | 30ms / 34fps | 23% | 20% | Save GPU for games | | CPU Low End | 720p, skip 3 | OpenCV SIMD | 27ms / 37fps | 21% | 20% | Older hardware |
Edge Refine toggle available for Killer and Zeus modes — adds ~27ms but recovers 89.9% of max quality edges.
Switch modes anytime from the Mode dropdown. No restart needed.
CUDA modes require the CUDA mode runtime: CuPy for compositing plus ONNX Runtime with
CUDAExecutionProviderfor model inference. Source,.deb,.rpm, and amd64 Snap installs handle this automatically on NVIDIA systems. The arm64 Snap build stays CPU-safe because ONNX Runtime GPU wheels are not published for Linux arm64 yet.
Architecture
NV Broadcast Pipeline
─────────────────────────────────
┌───────────┐ ┌─────────────────────────────────────────┐ ┌──────────────┐
│ Webcam ├────▶│ GStreamer Pipeline ├────▶│ Virtual Cam │
│(360p-4K) │ │ │ │ /dev/video10 │
└───────────┘ │ JPEG Decode ─▶ Color Convert ─▶ appsink │ └──────┬───────┘
└────────────────────┬────────────────────┘ │
│ ┌───────▼───────┐
┌───────────────▼────────────────┐ │ Chrome / Zoom │
│ Async Effects Thread │ │ Firefox / OBS │
│ (never blocks capture) │ │ Discord/Meet │
│ │ └───────────────┘
│ ┌──────────────────────────┐ │
│ │ AI Segmentation │ │
│ │ │ │
│ │ Pre-downsample to 720p │ │
│ │ (or 480/360 for Zeus/ │ │
│ │ Killer modes) │ │
│ │ │ │
│ │ ┌────┐ ┌─────┐ ┌────┐ │ │
│ │ │RVM │ │ISNet│ │BiR │ │ │
│ │ └──┬─┘ └──┬──┘ └─┬──┘ │ │
│ │ └──────┼──────┘ │ │
│ │ ▼ │ │
│ │ Alpha Refine │ │
│ │ (sigmoid+dilate) │ │
│ └────────────┬─────────────┘ │
│ │ │
│ ┌────────────▼─────────────┐ │
│ │ Edge Refiner (opt.) │ │
│ │ 720p 2nd pass RVM │ │
│ │ (Zeus/Killer only) │ │
│ └────────────┬─────────────┘ │
│ │ │
│ ┌────────────▼─────────────┐ │
│ │ Compositing │ │
│ │ │ │
│ │ ┌────────┐ ┌───────┐ │ │
│ │ │ Fused │ │ CuPy │ │ │
│ │ │ CUDA │ │ CUDA │ │ │
│ │ │ 0.1ms │ │ 15ms │ │ │
│ │ └────────┘ └───────┘ │ │
│ └────────────┬─────────────┘ │
│ │ │
│ ┌────────────▼─────────────┐ │
│ │ Video Enhancement │ │
│ │ 5 effects + presets │ │
│ │ GPU batch (CuPy) │ │
│ └────────────┬─────────────┘ │
│ ▼ │
│ Mirror flip (optional) │
└───────────────┬────────────────┘
│
┌───────────▼────────────┐
│ Preview (GTK4 Texture) │
│ Pause / Hide / Resize
Related Skills
python-debugpy
385.5kDebug Python with pdb, breakpoint(), post-mortem inspection, and debugpy remote attach.
skill-creator
385.5kCreate, edit, audit, tidy, validate, or restructure AgentSkills and SKILL.md files.
qqbot-channel
385.5kQQ channel management skill. Use qqbot_channel_api for explicit QQ channel-management requests; confirm write, delete, and bulk actions before calling authenticated QQ Open Platform endpoints.
claude-opus-4-5-migration
140.6kMigrate prompts and code from Claude Sonnet 4.0, Sonnet 4.5, or Opus 4.1 to Opus 4.5
