google translate 2
Image : BBC

Hola Google!

The recently released Google Translator app has taken language translation to a whole new level by adding its Voice Translation and Word Lens features. Voice translation allows users to speak in one language and computes the recorded voice to the target language by replying through a computer generated voice similar to Google Now.

The Word Lens app lets users take photos of sign boards and text written in a foreign language and translates the text into a native language based on user selection. This huge leap in computed translation sounds like a scene from science fiction movies but with Google’s plans about the technology, it won’t be too late when people all over the world could easily reduce the communication gap caused due to language variations.

Let’s explore in detail the two recent developments for Android and iOS users that hold an ocean of opportunities for both users and developers alike.

 See Also: Foreign Languages Can be Tough to Learn, Google Translate Makes It Easy

The Google Translate App

google-translate-app-logo

Google has worked on the translation technology since 2001 and its Google Translate web service has proven itself as a top translation technology available to users worldwide. More than 500 million monthly users translate text and information from over 90 available languages.

According to figures, the service handles over one billion translations every day and 95% of its users come from handheld or desktop platforms. This huge market has also led to a rapid development and accuracy of the self-learning technology that now stands as an unmatched translator.

Android supported real-time language translation since 2013 but this upgrade to the app eases the flow and immensely increases the accuracy in recognition and translation. The first version of the app could only translate one phrase at a time but now the app allows complete conversations to be translated among two speakers of different languages.

Google-Voice-Translate-Android
Voice Translation – As seen on the Android App

To use the app the user just needs to open the app and press the microphone button to allow the app to listen to you. If foreign language is detected, the app will compute the voice recording and translate it to the user’s native language. The app displays the translated text on-screen and speaks the phrase in a computer generated voice.

The user can also speak in his native language and the app will compute the recording to translate it to a target language. Once the first process is complete, the app will keep listening to further conversations and keep translating and speaking out in the 2 languages. This allows a seamless conversation with the device acting as an effective interpreter.

Word Lens

google translate 3
(Image : Google)

In its recent acquisition spree, Google acquired the Word lens technology from Quest Visual that developed the Word lens app for its target customers. After the acquisition, the app is inbuilt into Google Translator that allows its users to take pictures of sign boards, menus and other text written in foreign languages and displays its native translation on the device screen.

This technology can be highly beneficial to users travelling across foreign countries as the translation can easily help them gain a local insight and understanding without having to interact with the locals. In test runs, the app successfully translated information displayed on Pizza Menu’s written in supported foreign languages. The translation is also instantaneous allowing users to quickly take pictures and receive immediate translation of the text on screen.

Future Developments

Google aims to develop the app with some further additions that could ease user interaction and allow users to translate to even more languages in the near future. The Google Translate web service now supports more than 90 languages for text-based translations but the Android and iOS app only supports a few major languages for its translation based communication technology. Researches on all remaining languages are underway so that the next versions could support as many languages as possible.

The app works exactly like Google’s famous search engine. It crawls web pages and tracks translations among text to recognize the exact meaning of words and its contextual appearance. The algorithm determines the most feasible meaning and associates it with the word. Every time users use the word during their translation, the app learns its usage and applies it in future scenarios. This statistical analysis allows effective computation of languages to form necessary learning structures for the app.

Market Reactions and the Microsoft Factor

This upgrade to the app is a major threat for dictionary and translation technology based organizations that have struggled to gain a market share mainly dominated by giants like Google and Microsoft. In the recent upgrade to Skype, Microsoft released a live translation tool that aims to compete with Google. The upgrade allows users to effectively translate texts and speech between English and Spanish. The company aims to develop the technology to support 40 more languages in the near future.

Unlike Google that relies on its search based database of information, Microsoft relies on machine learning and only gets better with continuous usage. Therefore, the machine based learning algorithm forms a personalized database that varies from person to person. This database is unified by Microsoft to create a compilation of all possible languages algorithms that could in the future challenge Google’s translation technology.

On the other hand, Google relies on standard information and text available on the web to form its algorithmic structures. Therefore, none of the translations and results provided by Google is personalized for its target users. All information and translation are based on the majority usage on and available data on the web. Although information and translations are non-personalized, their usage patterns are standardized and used by majority of people all over the world.

Unfortunately, the lack of texts and information on some languages can hinder its effective algorithmic structure formation, which is not the case with Microsoft.

Conclusion

Translation technology has come a long way since their introduction by Google. Major third-party developers and dictionary software manufacturers have struggled hard to provide necessary usability to their customers through effective software. But due to the lack of technological expertise and information like Google and Microsoft, these companies often lag behind.

Google future additions and upgrades could greatly help its users communicate effectively with a wider audience while eliminating the need for a third person. But the technology is still in its infancy. There is a lot to be developed to replace the human interpreter.