Improving Accessibility With Speech-to-Text Technology at Minneapolis-St. Paul International Airport (MSP)
Client
Minneapolis–St. Paul International Airport (MSP)
System
Real-Time Gate Announcement Speech-to-Text
Category
Airports
Minneapolis–St. Paul International Airport (MSP): Real-Time Gate Announcement Speech-to-Text
Project Overview
As one of the busiest and most highly regarded airports in North America, Minneapolis–St. Paul International Airport (MSP) serves more than 36 million passengers annually. Consistently recognized for customer satisfaction, accessibility, and operational excellence, MSP is committed to ensuring every traveler can access critical information quickly and effectively.
To support that mission, MSP partnered with Electronic Design Company (EDC) to design and implement an innovative speech-to-text solution that transforms live gate announcements into real-time visual messaging, improving communication for all travelers while enhancing accessibility throughout the airport.
Problem
Terminal announcements—such as boarding calls, gate changes, and delay notices—rely heavily on audio broadcasts. However, spoken instructions exclude deaf and hard-of-hearing individuals, challenge non-native English speakers, and often disappear into ambient crowd noise.
MSP required a mechanism to convert live speech into readable text instantly. The project demanded software capable of parsing complex aviation terminology, differentiating multiple languages, and integrating with the existing paging infrastructure and third-party video displays without introducing latency.
Solution
EDC designed a cloud-connected integration bridging the physical gate microphones, the airport’s paging backbone, and the digital signage network.
Secure Audio Routing: The system intercepts live feeds from podium microphones and transmits the data to a cloud-based transcription engine.
Aviation-Trained Processing: The transcription software applies industry-specific vocabulary models to correctly format airline codes, gate numbers, and aviation terms.
Multi-Language Recognition: The engine identifies the speaker’s language and generates corresponding text to support diverse traveler demographics.
Synchronized Display: The processed text pushes to the specific gate’s digital screens in seconds, appearing alongside standard flight metrics.
Outcome
The speech-to-text deployment provides a visual counterpart to live gate announcements, fundamentally improving the travel experience for passengers who struggle with audio-only communications.
Unrestricted Accessibility: Deaf and hard-of-hearing individuals receive critical flight updates simultaneously with the rest of the terminal.
Lower Podium Congestion: Visual confirmation of boarding zones and seat assignments prevents travelers from crowding the gate desk to verify spoken instructions.
Seamless Expansion: Following successful initial deployments in Terminals 1 and 2, the established network architecture allows MSP to activate the technology at additional gates on demand.