Five Products · One Integrated Production Stack

Automating the Next Generation of Broadcast Media

Wyvern Tech engineers a connected real-time ecosystem of hardware and software platforms. We streamline virtual production, live translation, AI advertisement, broadcast automation, and computer-vision tracking into a single coordinated workflow.

Pioneering High-Efficiency, Single-Operator Production

Traditional live broadcasting requires fragmented rooms of separate hardware, complex coding matrices, and large crews. Wyvern Tech re-engineers this pipeline from the ground up, focusing on automation, latency reduction, and workflow integration.

Our connected stack operates on high-efficiency automation matrices, allowing a single operator to drive complex scene changes, live captions, camera telemetry, and ad injection. By simplifying operation and automating key technical tasks, we reduce project overhead while unlocking premium visual results.

0.02s
System Latency
100%
Pipeline Coverage
5-Inputs
Per VP Server
1-Op
Single Operator Workflow

The Product Ecosystem

Each product solves a specific layer of modern broadcast production. Together, they form a single coordinated workflow — from camera to graphics to translation to ad output.

Live Virtual Production Platform

Wyvern

A complete real-time broadcast studio on a single workstation. Wyvern integrates virtual production, green-screen compositing, and LED volumes into a live-switchable rendering pipeline driven by Unreal Engine. It allows a single operator to coordinate cameras, lights, and 3D scenes instantly.

  • Multi-Input Live Switcher: Up to 5 simultaneous inputs (SDI, NDI, ProRes, HAP) with frame-accurate transitions.
  • Composition & AR: Real-time chroma keyer with active reflections, shadows, and OpenColorIO color matrices.
  • Enhanced Shadows & Reflections: Skeleton-aware tracking generating virtual reflections on LED floors and cg-composites.
Wyvern Operator Interface showing Unreal Engine composition
Video Input
Up to 5x SDI / NDI / ProRes
Render Engine
Unreal Engine 5.8
Tracking Integration
FreeD, Stype, Mo-Sys, LiveLink
Composition Engine
AR, Green-Screen, & LED Composition in one Interface
Automation & Control
PTZ (Visca), Robotic Cranes, DMX, Art-Net, sACN, OSC
Operator Surface
Single Unified UI Console

Core Architecture & Modules

Wyvern re-engineers the broadcast control room into five unified software modules working inside a single server chassis.

Multi-Live Input Module Dashboard Preview

01Multi-Live Input

Accepts up to 5 simultaneous inputs (SDI, NDI, ST 2110) with per-input buffer controls. Tracking telemetry (FreeD, Stype, Mo-Sys) aligns to the video feed with sub-frame accuracy, adjusting focus, zoom, and phase offsets dynamically.

Chroma Keying LED Volume Compositing Preview

02Composition Matrix

Real-time green-screen keyer featuring edge extraction and color spill suppression. Integrates 3D virtual sets on LED volumes via nDisplay with full OpenColorIO calibration, generating reflections of actors in virtual glass and water.

Live Assets Library Interface Preview

03Live Assets Library

Every 3D asset, animation timeline, and data field runs live on air. Drag-and-drop workflow updates assets instantly without compiling. Features data-bound inputs connecting graphics directly to live election, weather, and sports feeds.

Operator UI Console Screen Preview

04Operator Experience

Consolidates launcher, editor, and live controls into one unified surface. Trigger macros handle camera movements, presets, graphics layers, and lighting (Art-Net / sACN / DMX) simultaneously in a single tap.

Enhanced Shadows & Reflections on LED volume floor

05Enhanced Shadows & Reflections

Skeleton-aware computer vision captures presenter coordinates on set. Feeds spatial data into render pipelines to project real-time shadows on LED floors, generate reflection buffers, and align tracking labels to the presenter.

Customer Onboarding & Partnership Workflow

Our four-stage onboarding pipeline delivers a ready-to-run show, backed by annual design, training, and live support.

Step 01
Physical Setup

Cameras, telemetry tracking nodes, LED panels, and studio lighting are physically installed and calibrated on site by our engineering team.

Step 02
Virtual Set Design

Our art department designs a fully customized, branded virtual environment in Unreal Engine, tailored specifically for your broadcast.

Step 03
Show Build

All scene layouts, trigger macros, data-bind scripts, and remote control widgets are authored and tested inside the Wyvern server.

Step 04
Handover

We hand over a ready-to-run live show. Your operator runs the daily broadcast easily via the unified surface with macros.

How Wyvern Compares

Capability Wyvern Regular VP Software
All-in-One Server Stack Switcher + Graphics + Automation Multi-system / Multi-box Modular
Live Inputs Per Server Up to 5 Live Cameras 1 Camera per server maximum
Render Engine Support Unreal Engine 5.8 Unreal Engine (Older)
Talent Shadows & Reflections Skeleton-Aware Tracking None or Partial (Static)
Operator Surface UI Single Unified Interface Multi-tool stacks (Director/Hub/Editor/Designer)
Daily News Operations By Design (Instant Cues/Macros) Slower to reconfigure / operation complexity
Preview

Voxa

A high-speed transcription and translation engine for live media feeds. Voxa takes in live audio streams, processes speech to text at sub-second speeds, and generates closed captions dynamically through a single-operator workflow, maintaining a record-low latency of around 0.02 seconds.

  • Real-Time Captioning: Automated speech-to-text with industry-specific vocabulary tuning.
  • Automated Pipeline: Streamlined translation output integrated directly into live broadcast feeds.
  • Broadcast Standard Outputs: Fully compliant with CEA-708, RTMP metadata inject, and web subtitles.
Voxa Live Captioning Interface Preview
Speech-to-Text Latency
~ 20 ms (0.02s)
Operator Requirement
Single Operator Control
Output Protocols
CEA-608/708, SRT, RTMP, WebVTT
Integration Method
REST API, Live Audio Feed, IP Stream
Vocab Customization
User-defined Dictionary Uploads
Security
Local Offline Server Deployment option
Preview

Apollo

An automated visual marketing generator. Apollo leverages neural model templates to ingest product briefs, asset coordinates, and brand files, immediately exporting consistent ad campaigns, broadcast commercials, and localized layouts with zero manual compositing.

  • Brief Ingestion: AI-driven context extraction from simple product images and descriptions.
  • Brand Consistency Guard: Automatic validation of color hex codes, logo safe spaces, and font alignments.
  • Multi-Format Output: Auto-renders crops for 16:9 widescreen, 9:16 vertical, and 1:1 ratios in parallel.
Apollo AI Advertisement Generation Preview
Generation Time
< 45 Seconds (Video Promo)
Aspect Ratios
16:9, 9:16, 1:1, Custom grids
Asset Formats
PNG, WebP, MP4, ProRes 422
Neural Engine
Wyvern AdDiffusion API
Brand Profiles
Up to 250 active profiles per system
Integration
Automated Ad Server feeds (AdManager)
Preview

Athena

The central operations hub for modern newsrooms. Athena connects the show rundown directly to robotic camera mounts, PTZ lenses, video switchers, graphics templates, and LED volume scenes. It coordinates a complex television lineup under a single, cue-based control surface.

  • MOS Newsroom Link: Real-time script and asset sync with AP ENPS, Octopus, and Inception.
  • Centralized Cues: Execute DMX lighting, camera tracks, graphics, and video triggers in one action.
  • Open API Architecture: Seamless integration with Ember+, REST, and WebSocket control interfaces.
Athena Broadcast Automation Rundown Preview
MOS Compliance
MOS Protocol v2.8.4 & v3.0
Control Protocols
Ember+, REST API, WebSockets, OSC
Robotic System Link
Panasonic, Vinten, Shotoku, Technodolly
Lighting Control
Art-Net, sACN, DMX-512 over Ethernet
Switching Latency
< 1 ms (Network Triggering)
User Interface
Web-based remote client (Responsive HTML5)
Preview

Vision

A cost-effective, high-precision camera tracking system. Vision uses optical markers and computer-vision pose estimation algorithms to calculate the exact spatial positioning and rotation parameters of camera rigs, streaming tracking telemetry directly over UDP.

  • Marker-Based Estimation: Real-time identification of ceiling/floor retroreflective visual markers.
  • High Frequency Solver: 120Hz pose estimates with sub-pixel lens distortion correction.
  • Direct Stream Out: Multicast UDP FreeD packets straight to Unreal nDisplay, AJA, and render nodes.
Vision Camera Tracking Estimation Solver Preview
Tracking Type
Retroreflective optical / active IR marker-based
Tracking Rate
120 frames per second
Latency
< 0.5 ms (Sub-frame pose update)
Stream Output
UDP multicast (FreeD protocol, XML coordinates)
Lens Calibration
Automatic radial/tangential distortion solving
Operating Range
Up to 25m ceiling heights, outdoor-capable

Expand Your Portfolio With Wyvern Systems

We work closely with global AV integrators, broadcast dealers, and studio engineering groups. As a Wyvern distributor, you gain access to tier-1 supply chains, volume pricing, hardware customization capabilities (OEM/ODM), and dedicated field engineering support.

  • Margin Protection

    Guaranteed dealer pricing tiers protecting your engineering margins.

  • Rapid Deployment

    Priority allocation on hardware batches with accelerated global shipping.

  • 24/7 Field Integration Support

    Direct priority line to our broadcast hardware support engineers.

B2B Portal Inquiry

Request technical brochures, distributor price cards, or arrange hardware demonstration slots.