Close
  • Latest News
  • Cybersecurity
  • Big Data and Analytics
  • Cloud
  • Mobile
  • Networking
  • Storage
  • Applications
  • IT Management
  • Small Business
  • Development
  • Database
  • Servers
  • Android
  • Apple
  • Innovation
  • Blogs
  • PC Hardware
  • Reviews
  • Search Engines
  • Virtualization
Read Down
Sign in
Close
Welcome!Log into your account
Forgot your password?
Read Down
Password recovery
Recover your password
Close
Search
Menu
eWEEK.com
Search
eWEEK.com
  • Latest News
  • Cybersecurity
  • Big Data and Analytics
  • Cloud
  • Mobile
  • Networking
  • Storage
  • Applications
  • IT Management
  • Small Business
  • Development
  • Database
  • Servers
  • Android
  • Apple
  • Innovation
  • Blogs
  • PC Hardware
  • Reviews
  • Search Engines
  • Virtualization
More
    Home Latest News
    • Mobile

    SwiftKey Mobile Predictive Keyboard Receives a Brainy Update

    By
    PEDRO HERNANDEZ
    -
    September 16, 2016
    Share
    Facebook
    Twitter
    Linkedin
      Microsoft SwiftKey

      SwiftKey, the mobile predictive keyboard specialist that Microsoft acquired earlier this year, has released an update to its eponymous app that promises to deliver smarter suggestions.

      While billed as a timesaver, the predictive text and autocorrect functionality built into the on-screen mobile keyboards can often become a source of frustration. Oftentimes failing to deliver the desired or appropriate suggestion, predictive keyboards can quickly turn a rushed text message or hurried email into meme fodder, or worse, spark a workplace misunderstanding.

      With its latest update to SwiftKey Keyboard for Android, based on neural-network technology that takes cues on how human brains process information, the company claims it has made a big leap in predictive keyboard accuracy.

      “With this update, SwiftKey is now able to meaningfully capture the relationship and similarity between words,” wrote the SwiftKey team in a blog post. “For example, having previously seen the phrase ‘Let’s meet at the airport,’ the technology is able to infer that ‘office’ or ‘hotel’ are similar words which could also be appropriate predictions in place of ‘airport.'”

      Previously, SwiftKey employed an “n-gram” model that analyzed the last couple of words in a message and offered suggestions based on words that commonly follow that sequence. In contrast, the new neural network-based system is trained with millions of sentences, enabling it to understand the relationships between words and predict phrases with better accuracy, even ones it never encountered during its training.

      For now, the updated system works in U.S. and UK English and is available as an update to the SwiftKey app in the Google Play store. Other languages are in the works, said Microsoft.

      The technology hails from the SwiftKey Neural Alpha project the group launched last fall, prior to being acquired by the Redmond, Wash., software giant. SwiftKey initially released the app last fall under its Greenhouse program. Similar to Microsoft Garage, Greenhouse is a collection of Android apps that allows participants to test SwiftKey’s latest innovations.

      Other projects include the Hexy Launcher, a home screen app that organizes commonly used apps into hexagonal tiles, and SwiftKey Symbols, which helps special-needs individuals communicate using symbol-based assistive technologies. Marking the 400th anniversary of the Bard of Avon’s death, in April SwiftKey released ShakeSpeak, a keyboard app that adds some Shakespearean flair to users’ writings.

      “SwiftKey’s technology learns your personal writing style to predict what you will most likely type next,” said Caroline Gasparin, SwiftKey’s head of Languages, in an April 6 announcement. “We have trawled thousands of words from his plays for SwiftKey to learn from the great playwright himself build this personalized language model.”

      In June, SwiftKey added a new feature, simply called Clipboard, that retains the snippets users copy or cut for an hour, in addition to the very last one, enabling them to paste items gathered during that time. Users can keep copied items longer by pinning them or add new items, like emails and other commonly used clippings.

      MOST POPULAR ARTICLES

      Android

      Samsung Galaxy XCover Pro: Durability for Tough...

      CHRIS PREIMESBERGER - December 5, 2020 0
      Have you ever dropped your phone, winced and felt the pain as it hit the sidewalk? Either the screen splintered like a windshield being...
      Read more
      Cloud

      Why Data Security Will Face Even Harsher...

      CHRIS PREIMESBERGER - December 1, 2020 0
      Who would know more about details of the hacking process than an actual former career hacker? And who wants to understand all they can...
      Read more
      Cybersecurity

      How Veritas Is Shining a Light Into...

      EWEEK EDITORS - September 25, 2020 0
      Protecting data has always been one of the most important tasks in all of IT, yet as more companies become data companies at the...
      Read more
      Big Data and Analytics

      How NVIDIA A100 Station Brings Data Center...

      ZEUS KERRAVALA - November 18, 2020 0
      There’s little debate that graphics processor unit manufacturer NVIDIA is the de facto standard when it comes to providing silicon to power machine learning...
      Read more
      Apple

      Why iPhone 12 Pro Makes Sense for...

      WAYNE RASH - November 26, 2020 0
      If you’ve been watching the Apple commercials for the past three weeks, you already know what the company thinks will happen if you buy...
      Read more
      eWeek


      Contact Us | About | Sitemap

      Facebook
      Linkedin
      RSS
      Twitter
      Youtube

      Property of TechnologyAdvice.
      Terms of Service | Privacy Notice | Advertise | California - Do Not Sell My Info

      © 2020 TechnologyAdvice. All Rights Reserved

      Advertiser Disclosure: Some of the products that appear on this site are from companies from which TechnologyAdvice receives compensation. This compensation may impact how and where products appear on this site including, for example, the order in which they appear. TechnologyAdvice does not include all companies or all types of products available in the marketplace.

      ×