Has Voice Recognition’s Time Come?

 - November 5, 2007, 6:26 AM

When Rockwell Collins introduced the Pro Line 21 integrated avionics system in 1996, the company proclaimed voice recognition would play a significant role in the avionics’ so-called man-machine interface. More than 10 years later the use of voice recognition in civil aviation has yet to emerge as a viable technology, but that could be about to change with the introduction of Pro Line Fusion.

Collins engineers are exploring a variety of uses for voice recognition, from moving the cursor to a particular window on a display to instructing the cursor to highlight a flight-plan leg and then execute a change. “Voice recognition holds a lot of promise,” said David Wu, director of flight deck systems marketing for Rockwell Collins. “The possibilities are pretty limitless, but we have to be careful that we first build up sufficient credibility with the regulatory agencies before we move into complex functions such as flight planning.”

He said engineers have defined some voice-recognition functions that perhaps seem mundane but which nonetheless will serve as a pathway for certifying advanced features. Simple voice commands that could be applied to Pro Line Fusion early on, for example, would include simple cursor positioning and radio tuning. Later, as the technology gains acceptance, pilots could use voice commands for a host of flight-related functions.

Although the use of voice recognition in civil aviation has yet to catch on, it is a common feature in the latest military cockpits. The Eurofighter and Joint Strike Fighter, for example, make heavy use of voice-recognition technology for command and control functions, including weapons selection. The pilot still must physically push the fire button, but he can use voice commands to choose the type of weapon that will do the job.

Experts agree that just about all fighters and attack helicopters in the future will incorporate some form of voice-recognition technology, and some say it is only a matter of time before similar concepts become commonplace in civil airplanes. In fact, Rockwell Collins has teamed with Adacel, the creator of the voice-recognition software in the Joint Strike Fighter, for the technology that will be applied in the Pro Line Fusion cockpit.

Lending credibility to the concept is the pace of advancements in the field in just
the last few years. Early voice-recognition software required a significant amount of training on the part of the computer software, which had to learn the nuances of a particular person’s speech pattern. Today’s software, however, is more sophisticated and adaptable. Not only can the latest voice-recognition systems understand almost any speaker, they are designed to decipher words even if the speaker’s first language is German, Italian, French or perhaps something else.

In fact, today’s software considers American English and British English two entirely different languages, and it is adept at understanding both.