Google Translate’s Camera Now Supports 60 More Languages
Traveling abroad this summer, and don’t speak the local language at your destination? The Google Translate mobile app may be a lifesaver.
Google this week updated the app’s instant camera translation feature, which lets you translate restaurant menus, signs, and even handwritten notes by simply pointing your camera lens at the foreign text. The update adds support for an additional 60 languages, including Arabic, Hindi, Malay, Thai, and Vietnamese.
With those additions, you can now translate from a total of 88 languages (check out the full list here) into more than 100. Previously, the app only translated other languages into English.
“This means you can now translate from Arabic to French, or from Japanese to Chinese, etc.,” Google Translate Product Manager Xinxing Gu wrote in a Wednesday blog post.
Besides that, the app can now automatically detect the language of the text you want to translate, a feature that may come in handy if you’re traveling in a region where people speak multiple languages. Instead of guessing, just select “Detect language” as the source language.
“Say you’re traveling through South America, where both Portuguese and Spanish are spoken, and you encounter a sign,” Gu wrote. “Translate app can now determine what language the sign is in, and then translate it for you into your language of choice.”
Google also gave the instant camera translation feature a bit of a makeover, which should make it “more intuitive to use.” On the bottom of the app, you’ll now see all three camera translation options: Instant, which translates text you point your camera lens at; Scan, which lets you take a photo then use your finger to highlight the text you want translated; and Import, which translates text from photos in your camera roll.
Finally, the feature should be more stable and work better overall, with less flickering and more accurate results.
“For the first time, Neural Machine Translation (NMT) technology is built into instant camera translations,” Gu wrote. “This produces more accurate and natural translations, reducing errors by 55 to 85 percent in certain language pairs.”