Google Translate App's Sign-Reading Capabilities Expanded - InformationWeek
IoT
IoT
Mobile // Mobile Applications
News
7/30/2015
08:06 AM
Connect Directly
Google+
LinkedIn
Twitter
RSS
E-Mail
50%
50%

Google Translate App's Sign-Reading Capabilities Expanded

Travelers can now translate a greater variety of languages in images more efficiently, even when offline. Foreign travelers in remote areas with poor connectivity now have broader communication options.

Windows 10: 10 Things To Know At Launch
Windows 10: 10 Things To Know At Launch
(Click image for larger view and slideshow.)

The Google Translate app has been accelerated and expanded to support realtime in-camera translation of 27 languages, up from seven in January, with the help of dirty letters and a self-contained neural network. The in-camera translations are handy when trying to ferret out the meaning of road signs or menu pages while traveling in a foreign country, for example.

Google Translate's text and speech translation capabilities still require a network connection to Google's data centers. But the app's longstanding ability to recognize and translate letters contained in images -- in realtime or in a snapshot -- while untethered from the cloud has become more efficient and more comprehensive. Foreign travelers in remote areas with poor connectivity now have broader communication options.

Google has managed this feat through sophisticated programming techniques and densely packed data. As Google software engineer Otavio Good explained in a blog post, the company created a small neural network inside the app.

"If you're translating a foreign menu or sign with the latest version of Google's Translate app, you're now using a deep neural net," said Good. "And the amazing part is it can all work on your phone, without an Internet connection."

Convolutional networks have proven to be effective for image recognition in recent years. But such networks often rely on large data sets stored in the cloud or local server banks. The CamFind app, for example, relies on neural networks and crowdsourcing for image recognition. But it requires a network connection to operate.

To create a compact neural network capable of recognizing the highly variable shapes of letters across multiple languages, Google engineers relied on sophisticated math and speedy SIMD processor instructions. To train the neural network, they created data sets using letters obscured by dirt.

"Letters out in the real world are marred by reflections, dirt, smudges, and all kinds of weirdness," explained Good. "So we built our letter generator to create all kinds of fake 'dirt' to convincingly mimic the noisiness of the real world -- fake reflections, fake smudges, fake weirdness all around."

(Image: Google)

(Image: Google)

Google introduced instant bidirectional image translation in Google Translate in January for English, French, German, Italian, Portuguese, Russian, and Spanish. Wednesday's update adds Bulgarian, Catalan, Croatian, Czech, Danish, Dutch, Filipino, Finnish, Hungarian, Indonesian, Lithuanian, Norwegian, Polish, Romanian, Slovak, Swedish, Turkish, and Ukrainian. In addition, the app can handle one-way image translation from English to Hindi or Thai. The app supports these 27 languages for realtime image translation, and 37 languages using stored snapshots.

Google previously made image recognition available in its Google Goggles app for Android in December 2009. Google Goggles was subsequently released for iOS and then discontinued for iOS in 2014. Around the same time, Google acquired Word Lens, an image recognition and translation app, that was folded into Google Translate in January.

Thomas Claburn has been writing about business and technology since 1996, for publications such as New Architect, PC Computing, InformationWeek, Salon, Wired, and Ziff Davis Smart Business. Before that, he worked in film and television, having earned a not particularly useful ... View Full Bio

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Angelfuego
50%
50%
Angelfuego,
User Rank: Ninja
7/31/2015 | 6:49:10 PM
Re: Hepful provided it works successfully
@Sunita, So true. This app capability will definitely come in handy when traveling. I sure wish I had this on my last trip. It would have made things a lot easier for me.
Angelfuego
50%
50%
Angelfuego,
User Rank: Ninja
7/31/2015 | 6:47:42 PM
Re: Hepful provided it works successfully
@jagibbons: I also have had a positive experience with Google Maps and Navigation. I have never encountered a problem with it. It is a quick and convenient navigation tool that works well for me.
Angelfuego
50%
50%
Angelfuego,
User Rank: Ninja
7/31/2015 | 6:45:40 PM
Expanded Capabilities
I am excited to try Google Translate's expanded capailities. It will come in handy with my line of work. I can't wait to try it.
Angelfuego
50%
50%
Angelfuego,
User Rank: Ninja
7/31/2015 | 6:45:37 PM
Expanded Capabilities
I am excited to try Google Translate's expanded capailities. It will come in handy with my line of work. I can't wait to try it.
jagibbons
50%
50%
jagibbons,
User Rank: Ninja
7/31/2015 | 3:23:59 PM
Re: Hepful provided it works successfully
That has never been my experience with Google Maps and Navigation. It had always worked well for me.
SunitaT0
50%
50%
SunitaT0,
User Rank: Ninja
7/31/2015 | 2:40:33 PM
Re: Hepful provided it works successfully
Google Maps is a terrible guide. It is almost never right and its gets too busy in updating the current location, after you have passed by it.
SunitaT0
50%
50%
SunitaT0,
User Rank: Ninja
7/31/2015 | 2:37:55 PM
Re: Hepful provided it works successfully
If only this arrived sooner when I visited Japan. I got so lost, English signs were extremely rare in Tokyo, and people weren't willing to talk too much. This app would make travelling so easy. Because then you wouldn't spend time trying to understand or read the local language.
jagibbons
50%
50%
jagibbons,
User Rank: Ninja
7/30/2015 | 12:56:19 PM
Re: Hepful provided it works successfully
It's not going to translate everything using the densly packed offline data, but it should be able to do quite well on standard road signs and other common markers. Especially in transportation, a limited vocabulary is used intentionally to reduce barriers to understanding and safety.
asksqn
50%
50%
asksqn,
User Rank: Ninja
7/30/2015 | 12:41:33 PM
Hepful provided it works successfully
Provided that the app successfully translates into the user's intended language, Translate seems like a tremendously useful tool for travelers.  On the other hand, if Translate works along the lines of Google Maps, then the user will, no doubt, run into similar problems maps users do, i.e. uknown construction and detours.
Commentary
AI & Machine Learning: An Enterprise Guide
James M. Connolly, Executive Managing Editor, InformationWeekEditor in Chief,  9/27/2018
Commentary
How to Retain Your Best IT Workers
John Edwards, Technology Journalist & Author,  9/26/2018
Slideshows
10 Highest-Paying IT Job Skills
Cynthia Harvey, Contributor, NetworkComputing,  9/12/2018
Register for InformationWeek Newsletters
Video
Current Issue
White Papers
Slideshows
Twitter Feed
Sponsored Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.
Sponsored Video
Flash Poll