The coolest thing about wearable technology is the fact that so many of the devices we’ve seen in movies, televisions and comic books are coming to life, such as the Star Trek Communicator – a voice activated device that enabled a wide range of actions.
Voice recognition technology is nothing new – having been around since the 1950s when introduced by Bell Labs. However, it wasn’t until the 1990s when IVR systems (interactive voice response) systems became more widespread that the technology really began to solidify.
Unfortunately, many of these IVR systems are not recollected with much fondness as they historically have been plagued with low accuracy levels – and I am sure that I am not alone in having resorted to name-calling when using one of these systems.
Thankfully, two product introductions have done more to move voice recognition forward than Star Trek. The first was the introduction of Google Voice Search which lets users ask questions out loud and get answers spoken back. It currently incorporates 230 billion works from actual user queries – but only for US English only. The second is Apple’s Siri – which really needs no explanation – except to say that it only works with iOS operating systems, but does offer multiple languages. These combined with the rapid adoption of smartphones has helped to push voice as a user interface into the mainstream.
Why Voice makes Sense as a User Interface
As consumers grow more comfortable with voice recognition technology for everyday tasks, developers are putting speech into a variety of devices including remote controls, cars and wearable technology.
Voice as a user interface makes sense because it is something that comes naturally to humans and is used on a daily basis. According to research conducted by Matthias Mehl, the average number of words spoken by an adult per day is 16,000, while the average number of words typed per day is 3,000 to 4,000.
A key advantage is the fact that voice is not only hands-free, but eyes-free, making it suitable for use in a variety of environments, as well as for the visually-disabled. And unlike most written words, voice can communicate mood, gender, identity (such as recognizing a voice), emphasis and even personality.
But perhaps most important, is the fact that the user does not need to learn how to use a new application – they simply speak what they want to do.
Despite all of these advantages, using voice as the user interface is not without its challenges – particularly for developers.
The Challenges of Using Voice User Interfaces
According to the book “Don’t Make Me Tap” written by XOWi founders Ahmed Bouzid and Weiye Ma, there are four key dimensions that must be considered when using voice as the user interface and to deliver a good user experience:
- Environment – where is this interaction to take place and are you in the company of other people
- Content – what and how much is being communicated
- User State – what is the state of the user
- Medium – how should the interaction be carried out?
For instance, in a noisy environment it may make sense to deliver the information via text rather than voice. Or in a public environment, it may not make sense for the user to disclose credit card information.
Other considerations for implementing a voice user interface include understanding the user needs and intent, incorporating support for language nuances and possibly slang terms. I always like to reference the differences in U.K English versus U.S. English in a few of my favorites:
Interesting Applications for Voice UIs
There are a number of notable introductions of products and applications driven largely by using voice as a user interface – most notable, of course is Google Glass, but we are also seeing widespread adoption by the automotive industry.
According to IMS Research, 55 percent of all new motor vehicle vehicles produced in 2019 will incorporate voice recognition, up from 37 percent in 2012. Voice recognition technology is currently being deployed by a number of car manufacturers to allow drivers to use spoken words to manipulate an increasing number of functions, including controlling the HVAC system, entertainment system, sending text messages and even composing email.
Another interesting product is the XOWi Personal Voice Companion badge - a small wearable device that can be clipped on or worn as a necklace. It provides a completely hand-free/eyes-free experience by using voice interactions to accomplish numerous tasks such as control your AC, get traffic and weather conditions (to name only a small few). It pairs with a user's smartphone via a companion app that runs on multiple mobile platforms. It also offers an open application programming interface that will enable developers to add functionality and content to XOWi.
As a fitness enthusiast – I would love a fitness device that would enable me to simply ask – how far have I gone? What is my heart rate? How many calories have I burned? Rather than scrolling through menus or frankly even looking at the device – particularly when I'm riding my bike. Developers – are you listening? And if this exists – please let me know!
We are already seeing a number of applications that use your voice as your password. The movie “Sneakers” – made this line famous with its “verify me” statement. But a number of smartphones – such as Moto X are widely incorporating this technology via Google Now to deliver a “touchless” experience.
We talked about the 4W’s of wearable tech in a previous blog and concluded that the key to wearable tech will be to design a product that fits seamlessly and unobtrusively into a persons’ life. We also concluded that the biggest challenge will be to keep the user engaged and continually using the device. Voice is the most natural and easiest fit to keep consumers engaged and will become a necessity as most wearable devices will be in increasingly smaller form factors.
So, beam me up Scotty!