Machine learning is accomplishing Facebook's mission of connecting the world across language barriers. Facebook is now serving 2 billion text translations per day. Facebook can translate across 40 languages in 1,800 directions, like French to English. And 800 million users, almost half of all Facebook users, see translations each month. That's all based on Facebook's own machine learning translation system.

Professor Mirella Lapata has received five years' funding for her project, TransModal: Translating from Multiple Modalities into Text. The European Research Council (ERC) Consolidator Grant worth 1.9M will begin in September. ERC Consolidator Grants are for researchers of any nationality with 7-12 years of experience since completion of their PhD (plus 18 months for each child), a scientific track record showing scientific talent and an excellent research proposal. Professor Lapata's award winning proposal is summarised on the ERC website as follows: "Recent years have witnessed the development of a wide range of computational methods and tools that process and generate natural language text. Many of these have become familiar to mainstream computer users such as tools that retrieve documents matching a query, perform sentiment analysis, and translate between languages. Indeed, publicly available systems like Google Translate can instantly translate between any pair of over fifty human languages allowing users to access web content that wouldn't have otherwise been available. "The accessibility of the web could be further enhanced with applications that not only translate between different languages (eg.

Machine learning is accomplishing Facebook's mission of connecting the world across language barriers. Facebook is now serving 2 billion text translations per day. Facebook can translate across 40 languages in 1,800 directions, like French to English. And 800 million users, almost half of all Facebook users, see translations each month. That's all based on Facebook's own machine learning translation system.

Machine learning is accomplishing Facebook's mission of connecting the environment throughout language obstacles. Fb is now serving 2 billion textual content translations for every working day. Fb can translate throughout 40 different languages in 1800 instructions like French to English. And 800 million people, pretty much 50 % of all Fb people, see translations just about every month. Alan Packer, Facebook's Director of Engineering for language technological innovation, discovered this progress nowadays at MIT's Emtech Electronic convention in San Francisco.

Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural machine translation often belong to a family of encoder-decoders and consists of an encoder that encodes a source sentence into a fixed-length vector from which a decoder generates a translation. In this paper, we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder-decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word, without having to form these parts as a hard segment explicitly. With this new approach, we achieve a translation performance comparable to the existing state-of-the-art phrase-based system on the task of English-to-French translation. Furthermore, qualitative analysis reveals that the (soft-)alignments found by the model agree well with our intuition.

Need a translation app for your vacation but don't want to get slammed with a big data bill? New updates to Google Translate, including the addition of another major language and a pop-up translation feature, can help. Google said today that its translation app -- which can translate in a number of ways including hearing someone speak or by reading what they write -- can now be used in offline mode with no data or Wi-Fi connection on both iOS and Android (it previously wasn't available on iOS), potentially eliminating the high price of data for travelers with iPhones. Some 52 languages, including French, German, Russian and, most recently, Filipino, can be translated offline. A complete list is here.

If trying to order dinner or find your hotel abroad fills you with fear due to your abysmal grasp of foreign languages, don't panic. A forthcoming in-ear gadget is claimed to be able to translate speech like the Babel Fish in The Hitchhiker's Guide to the Galaxy, or the Universal Translator gadget in Star Trek. The system, dubbed the Pilot, comprises two earpieces to be worn by two people who do not speak the same language and uses an app so the duo can converse with ease. The Pilot system comprises two earpieces (shown in three colours above) to be worn by two people who don't speak the same language It is claimed to be the first'smart earpiece' capable of translating between two languages. The company behind the technology, Waverly Labs, said: 'This little wearable uses translation technology to allow two people to speak different languages but still clearly understand each other.'

Recent advances in conditional recurrent language modelling have mainly focused on network architectures (e.g., attention mechanism), learning algorithms (e.g., scheduled sampling and sequence-level training) and novel applications (e.g., image/video description generation, speech recognition, etc.) On the other hand, we notice that decoding algorithms/strategies have not been investigated as much, and it has become standard to use greedy or beam search. In this paper, we propose a novel decoding strategy motivated by an earlier observation that nonlinear hidden layers of a deep neural network stretch the data manifold. The proposed strategy is embarrassingly parallelizable without any communication overhead, while improving an existing decoding algorithm. We extensively evaluate it with attention-based neural machine translation on the task of En->Cz translation.

Translating text on your smartphone just got a whole lot easier. Google on Wednesday rolled out new updates for the Android and iOS versions of its Translate app. Android users are getting a new feature called Tap to Translate, which lets you translate text just by highlighting it. The feature works in any app that lets you highlight, according to Google. When you select some text, a Translate icon will appear at the top of your screen.

Translate for iOS now includes offline support, giving you a way to communicate in other languages when you don't have data service (say, on vacation). And if you regularly visit China, you'll be glad to know that camera-based Word Lens translation on both Android and iOS now supports simplified and traditional Chinese. If you've ever struggled to make sense of a Beijing restaurant menu or a Shanghai street sign, you can rest easy.