API Build-data JSON Resources
Theme

Choose how MSRBot.io looks on this device.

Preference is stored in this browser only.

MTS 2025, Article 9 (pp. 1 to 11)
[ACTIVE]

VoxiumAI

Metadata

Publisher
SMPTE — Pasadena, CA
Doc Type
Conference Paper
Content Type
Original Research
Volume
00, pp. 1–11
Abstract
VoxiumAI: Voice-Activated Control for Live Production This paper presents VoxiumAI™, a production-ready voice control platform that enables live broadcast systems to respond to natural speech in real time. Designed to simplify complex control workflows in high-pressure environments, VoxiumAI uses automatic speech recognition (ASR), intent parsing, and a context-aware state machine to interpret spoken commands and dispatch control signals to production hardware such as graphics engines, video switchers, and automation systems. VoxiumAI was deployed in a major 2024 election broadcast, where on-air talent used natural language to trigger immersive augmented reality graphics without operator intervention. Commands such as “let's take a look at Fulton County” or “let's clear this…” were disambiguated and interpreted live, using a workflow engine that linked real-time ASR with a state machine aware of the current visual context. This allowed the system to distinguish between similarly named geographic regions and deliver precise, seamless on-air transitions. Additional integration with tracking hardware enabled interactive 3D storytelling inside a fully virtual environment rendered in real time. The deployment demonstrated the potential of AI-enhanced voice workflows to increase responsiveness, reduce control room burden, and expand creative flexibility for producers and talent. This paper describes the system architecture, implementation challenges, and lessons learned from the live event deployment. It also outlines the broader implications of voice-driven automation for news, sports, and live events, offering a path forward for integrating speech interfaces into future broadcast workflows.
Publication Date
2025-10-13
ISBN
[object Object]
Author(s)
Jim Doyle
Keyword(s)
Voice Control, Artificial Intelligence, Speech-to-text, Contextualizer, Context Engine, RossTalk, Broadcast Graphics, Augmented Reality, Live Production, Election Coverage, Workflow, Video Switchers, Control Room Automation, Sports Broadcasting, Esports Broadcasting, Corporate AV, Production Workflow
Copyright
© 2025 SMPTE
Source Data (JSON)

Full registry record with provenance metadata. Open directly: /api/doc/978-1-61482-966-9-9.json

Reference this Doc

Plain text (ISO 690 compliant)

Preview:
Jim Doyle; VoxiumAI, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]
Snippet:
Jim Doyle; VoxiumAI, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]

HTML (ISO 690 compliant)

Preview:
Jim Doyle; VoxiumAI, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]
Snippet:
<span class="citation">Jim Doyle; <cite>VoxiumAI</cite>, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]</span>

SMPTE Icon SMPTE's HTML Pub

Preview:
Jim Doyle; VoxiumAI, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]
Snippet:
<li>
Jim Doyle; <cite id="bib-978-1-61482-966-9-9">VoxiumAI</cite>, MTS 2025, Article 9 (pp. 1 to 11); SMPTE, 2025, ISBN: [object Object]
</li>