Touch-typing keyboard apps for mobile devices flood Mobile World Congress

BARCELONA—The average person hunts and pecks entire sentences on a smartphone keyboard multiple times a day, whether it's a quick text message, a longer email, or an entire article or document.

With most smartphones lacking a physical keyboard, however, that means endlessly moving fingers across a flat screen with little to no capacitive response. This isn’t the most ergonomic approach, and our hands have to adapt to the motion and method.

Although features such as predictive text, autocorrect, and alternative typing methods like Swype are designed to help people type more efficiently, they aren’t the best option for every user.

To try to fix this problem, app developers are jumping into smartphone keyboards, and I saw quite a few of these types of apps here this week. Each app has its own innovative method for promoting better typing, but the general message is loud and clear: There has to be a better way to type on a touchscreen.

One step ahead

Perhaps more-advanced predictions is the way to go. That’s what the team behind SwiftKey has been working on since the first Android handset launched in 2008.

SwiftKey developers created a stand-alone Android app that learns common phrases and words from you as you type. Over time, the app understands the context of language and suggests words for you before you even start to enter them.

The app connects with your Facebook page, Gmail account, Twitter handle, RSS feed, or blog to learn more about you and the language you naturally use. The longer you use SwiftKey, the more it understands, and the less work you have to put into typing.

Here's a video showing how SwiftKey works:

“Autocorrect isn’t thinking about sentence context and what comes next,” says Joe Braidwood, SwiftKey’s chief marketing officer. “So we thought, ‘That’s dumb, we’re going to change this.’”

Braidwood explains that the original Android keyboard suffered in earlier versions due to the platform's rushed launch to compete with the iPhone. That gave SwiftKey an advantage, and the app became widely popular in the Google Play Store when it launched in 2010. By 2012, it was the best selling paid Android app globally.

If you look at the evolution of the built-in Android keyboard, its predictive capabilities are improving, but they still seem to be one step behind SwiftKey’s technology.

As SwiftKey continues to distinguish itself as a leader in predictive text—the company announced continued growth with its new SwiftKey Healthcare platform at MWC—other developers are starting to follow suit.

The team at WordLogic is taking a similar approach: They have built a solid prediction technology through different dictionaries that can work in stand-alone apps.

WordLogic has been in the predictive-text game since the late 1990s, having produced the keyboard tech behind the Palm Pilot (remember those?), but the company was at MWC this week to promote its new keyboard app for Android, iKnowU.

Like SwiftKey, iKnowU learns from the user and remembers common phrases and words. However, iKnowU excels in predicting a string of words together, such as a phrase or a short sentence. It also guesses which keys you’re likely to press next, which it highlights on the screen in a bright green or blue. The colors represent how many predicted words are associated with that key.

Here's iKnowU in action:

If you want iKnowU to remember some slang, names, or other unusual words that you use often—without waiting for it to actually, well, get to know you—you can enter them manually. This combination of predictions and stored phrases makes typing on this keyboard lightning fast once you’ve had some practice.

Because Apple doesn’t allow additional keyboard apps on its devices, the company behind SwiftKey has entered partnerships with other developers to embed this technology in apps for iOS; WordLogic is likely to do the same down the road.

Next: Snapkeys Si and TouchPal Keyboard

For comprehensive coverage of the Android ecosystem, visit Greenbot.com.

Subscribe to the Best of TechHive Newsletter

Comments