Getting the Message

AI-powered speech-to-text may reshape how passengers experience communication at 35,000 feet.

September 05, 2025 in Technology, Innovation

airplane seat back screen shows welcome message WELCOME WITH WORDS: A mock-up demonstrates Boeing’s developmental speech-to-text transcription concept. The prototype shows cabin announcement captions in nearly real time to enhance accessibility for passengers. IMAGE: ©BOEING

When people board an airplane, they enter an environment where the passenger experience has been carefully engineered. But for millions of travelers who are Deaf or Hard of Hearing (DHH), one vital part of that experience — spoken communication — is missing. Meal service options, turbulence notices, gate changes — all are moments when hearing passengers are informed, and others are isolated.

At the Aircraft Interiors Expo (AIX) 2025 in Hamburg, Germany, Boeing demonstrated a speech-to-text transcription (STT) concept that displays captions of cabin announcements in nearly real time.

three engineers sit in demo area TEXT TEST: Boeing engineers, from left, Hazel Nam, Jinri Kim and Jihyun Kim, check the STT prototype in a test environment. PHOTO: ©BOEING

Driven by artificial intelligence (AI) and designed to function offline, the developing technology could bridge a long-standing accessibility gap and make air travel more inclusive.

“With each accessibility advancement, our global team is building toward a more inclusive travel experience for everyone,” said Boeing engineer Ashley Badger.

Inclusive design

“The real problem wasn’t technology,” said Bill Harkness, Boeing accessibility engineering leader. “It was access. It was dignity.”

Harkness, who is Deaf, has long advocated for better communication tools for passengers who, like him, rely on visual information.

“This was our chance to change that,” he said. “To build a system that doesn’t assume one way of listening fits all.”

engineer addresses group of people ACCESSIBLE INNOVATION: Bill Harkness, left, is guiding the development of Boeing’s STT concept. PHOTO: COURTESY OF BILL HARKNESS

The earliest ideation didn’t start with code, but with a question: How could technology make cabin communication more inclusive? The work reflects a growing shift toward more human-centered, inclusive cabin technologies and prototyping to demonstrate concepts early.

“We needed to understand the user’s perspective before diving into the technology,” said Amy Goodell, principal investigator. “You have to identify the right problems in order to come up with the right solutions.”

Above the noise

The STT concept emerged as Boeing engineers and research scientists began to collaborate.

Jinri Kim, a machine learning researcher at the Boeing Korea Engineering & Technology Center (BKETC), recalls how a simple idea gained momentum.

“Developing the prototype, we had to select a model that could understand context — what’s a crew announcement and what isn’t,” said Kim.

Aircraft cabins present serious technical challenges, including persistent engine noise, overlapping audio sources and limited Wi-Fi connectivity. To meet those constraints, the BKETC AI team evaluated multiple open-source architectures and adapted a model to work in noisy, offline conditions. The system processes incoming audio in two-second chunks, allowing near-real-time captions to appear on local devices or embedded displays.

“We had to avoid hallucination,” said Kim, referring to how AI sometimes generates plausible but incorrect output. “In-flight systems don’t have the luxury of guessing.”

“We selected models that could handle noisy input,” explained Jihyun Kim, a BKETC data scientist. “You can’t assume a clean signal on board.”

three tablets display message PERSONAL TOUCH: Personal tablets show the STT concept at AIX 2025. The system is designed to run offline and adapt to noisy cabin conditions. PHOTO: ©BOEING

Hazel Nam, a BKETC machine learning researcher, added, “It wasn’t just about accuracy in the lab. We wanted to know if people in a cabin would understand the message when it mattered.”

As the teams developed the prototype, the project followed a human-in-the-loop approach — anchored not just in machine learning, but in human insight and accountability.

Demonstrations of the prototype are showing promise, according to Boeing engineer Carlota Ramiro De Huelbes, who is working with the BKETC AI team to mature the prototype and deploy it in a more realistic cabin environment.

The current prototype runs offline and has been tested on portable tablets in a mock-up. Teams are designing the technology for integration on both legacy and next-generation aircraft systems and are exploring additional features, such as language translation.

Nothing about us, without us

Most importantly, the STT concept is being developed hand in hand with Deaf engineers and community members — not just for them.

“We call it co-design,” said Harkness. “But really, it’s shared authorship.”

Deaf and Hard-of-Hearing visitors at AIX responded with visible emotion, with many noting it was the first time they had seen themselves served by innovation in airplane cabins.

woman speaks into microphone and second woman uses keyboard ONE MESSAGE AT A TIME: Amy Goodell and Carlota Ramiro De Huelbes test and modify the prototype code to appear on portable tablet displays. PHOTO: ©BOEING

While the system was initially developed with DHH passengers in mind, the benefits may extend to all travelers — including anyone who has struggled to hear announcements over engine noise or during busy service times. It’s a reminder of a key principle in inclusive design: Accessibility helps everyone.

The goal remains simple: Make flying more inclusive — one message at a time.

By Eunnyoung Min