[Avionics Today 11-13-2014] In the future, pilots will be able to use voice recognition technology to perform flight deck tasks and functions that today are only manageable by hand. While cockpit voice recognition technology has been in development for over a decade, engineers at the Rockwell Collins Advanced Technology Center are progressing to the point where the capability is no longer a question of how, but when.
Rockwell Collins’ Pro Line Fusion flight deck. Photo: Rockwell Collins.
To demonstrate how close the technology is to becoming a reality for pilots, Geoff Shapiro, a human factors engineer for Rockwell Collins, gave Avionics Magazine a live look in to the Advanced Technology Center using a webcast service, demonstrating how the sophisticated voice recognition system works. The voice recognition system is controlled by pressing a button on an airplane control stick and providing voice commands. After the pilot gives the voice command, the system’s synthetic voice engine repeats the command and then shows it on the active cockpit display where a pilot can either accept it if it is correct, or reject it and repeat the command until the system gets it right.
“Our flight decks have more information access than they’ve ever had before, which is really fantastic, but with that information access there comes a need to be able to get to it quickly, so we’re investigating any technology that would help control the flight deck and get the information quickly and rapidly,” Shapiro said. “Speech recognition is one of those technologies that is really neat because it does a couple things. One, it allows me to control the flight deck without actually controlling any of the technologies in the flight deck so that means I don’t have to look down, I continue looking out the window, which is really important to keep your eyes focused on the mission at hand and minimize the amount of time that you’re sticking your head down and messing with knobs and buttons.”
Rockwell Collins regularly invites pilots into the Advanced Technology Center to test the technology, so they can find flaws and make improvements. One of the observations they’ve made recently is that the voice recognition technology allows pilots to reduce the time required to perform complicated tasks.
For example, it typically takes a pilot 30 seconds to perform commands received from Air Traffic Controllers (ATCs), such as instructions regarding setting the heading, altitude and radio frequency changes among other tasks. The voice recognition technology reduces that time to eight seconds.
“Being able to control your FMS [Flight Management System], that’s huge. They’re very excited about those kinds of functions. When the workload is really bad, that’s when this technology is really going to help them,” Shapiro said referring to how the technology can help pilots. “We don’t anticipate replacing any flight deck control panel with speech recognition, it is aimed more at assisting with reducing the pilot workload.”
Shapiro also pointed out that voice recognition technology is already being used for other modes of transportation, such as voice activated GPS systems used in automobiles. There are also examples today of voice recognition being used in military aircraft. The Eurofighter Typhoon has a direct input voice system that recognizes more than 90 possible voice commands. Lockheed Martin has also designed the F-35 cockpit with a speech recognition system. Airbus Defense and Space is also considering adding voice recognition technology to its newly developed Sferion assistance system for helicopter cockpits.
However, deploying voice recognition technology in the business jet and eventually the commercial air transportation world will take some time.
“Its not a technology problem really there its more of an acceptance problem, air transport tends to be a little bit more lagged compared to the business jet market. Once they see its actually out there and proved itself and it has proved its value then air transport will accept it,” said Shapiro.
Currently, the Advanced Technology Center engineers are examining how the technology would best assist a flight crew in an air transport aircraft versus a smaller aircraft operated by a single pilot.
“If you have a flight deck operated by two pilots with the communication between the other seat, yourself and Air Traffic Control, we think maybe the synthetic speech might be audio overload,” said Shapiro. “Jumping down to a King Air aircraft where there’s only one pilot or a rotary wing operation, that may still offer a benefit because there’s less audio chatter in the flight deck and adding that synthetic voice option back in will lead to less interference.”
While Shapiro did not offer a timeline for actual deployment and regular usage of the technology by pilots, he says his engineering team is getting close.
“Its only recently we’ve gotten to the point where our processing capability has improved to the point where its able to work with the speech recognition engine and speech recognition,” said Shapiro. “We’ve hit our sweet spot finally and its gotten to the point where its getting very, very close to being product ready in terms of being mature enough to get out there.”