The results obtained from Neural Machine Translation are amazing, in particular, the neural network’s paraphrasing. It almost seems as if the neural network really “understands” the sentence to translate. In this first article, we are interested in “meaning,” that which gives an idea of the type of semantic knowledge the neural networks use to translate.
Let us start with a glimpse of how the 3 technologies work, the different steps of each translation process and the resources that each technology uses to translate. Then we will take a look at a few examples and compare what each technology must do to translate them correctly.
SYSTRAN, a global leader in language translation technology, will demo two new integrations for SYSTRAN’s offering of Relativity at Relativity Fest 2016 this Sunday in Chicago.
SYSTRAN is attending Relativity Fest as a silver sponsor. The company plans to demo two new integrations, aDiscovery and Anonymizer, for SYSTRAN’s offering of Relativity at the event. The aDiscovery feature will aid in audio discovery by transcribing audio files, detecting the source language and then translating the content. Anonymizer applies rigorous anonymization techniques to the full text and meta data of electronic documents within Relativity.
A Deep Dive into SYSTRAN’s Neural Machine Translation (NMT) Technology
by Kirti Vashee on eMpTy Pages, a blog about translation technology, localization and collaboration
[…] So, I recently had a conversation with Jean Senellart , Global CTO and SYSTRAN SAS Director General, to find out more about their new NMT technology. He was very forthcoming, and responded to all my questions with useful details, anecdotes and enthusiasm. The conversation only reinforced in my mind that “real MT system development” is something best left to experts, and not something that even large LSPs should dabble with. The reality and complexity of NMT development pushes the limits of MT even further away from the DIY mirage.
In the text below, I have put quotes around everything that I have gotten directly from SYSTRAN material or from Jean Senellart (JAS) to make it clear that I am not interpreting. I have done some minor editing to facilitate readability and “English flow” and added comments in italics within his quotes where this is done.
Project “PNMT” for Purely Neural Machine Translation was this year’s flagship project for the researchers and developers at SYSTRAN.
SYSTRAN brings its expertise in several ways: contributing to research on neural models; applying its know-how in terminology to increase the potential of Neural Machine Translation; and industrializing technology to make it available to companies, organizations and individuals.
We will keep you posted each month and share best practices, research paper, customers insights, product news…
The first engine to be based on neural models and deep learning, delivering unparalleled translation quality!
Project “PNMT” (Purely Neural Machine Translation) was this year’s flagship project for the researchers and developers at SYSTRAN, the leading provider in machine translation and natural language processing, confirming its pioneer position for over 40 years.
SYSTRAN brings its expertise to the sector in several ways: contributing to research on neural models; applying its know-how in terminology to increase the potential of Neural Machine Translation; and industrializing technology to make it available to companies, organizations and individuals.
In 1968, during the height of the Cold War, our nation’s intelligence organizations and military were working to decipher and understand Russian communication threads coming through their offices. Answering the call, SYSTRAN was born. We provided the vital translation software that our nation’s security forces required in order to do their job effectively.
Back then, we were providing our services through the IBM mainframe. Thankfully technology got smaller and more advanced, allowing SYSTRAN to move into several different areas. With the advent of the PC, we developed a product that people can use on the desktop, translating text into 130+ language combinations, enabling real-time multilingual communication across the globe.
Then recently, just in the last decade, we experienced further advancements through the smartphone. Speech recognition software along with the smartphone provided a plethora of breakthroughs. For example, SYSTRAN developed S-Translator, the official translation solution on the Samsung Galaxy S and Note series. Now if you’re in a taxi in Korea you can use your phone to tell someone where to go without speaking a word of the native language.
“Languages are intriguing and challenging at the same time”
As kids we were always intrigued by the way Google Translator worked. While it translated those famous French quotes for us, there were limitations which even Google couldn’t surpass. Since then, language has been a barrier— hindering our global crusades. Be it a worldwide competition or business meetups across countries, a common language would have been the best idea which sadly is pretty hard thing to materialize.
Even readers at the Huffingtonpost must have had difficulties with other country specific domains, offering great pieces of work which couldn’t be accessed— owing the language barrier.
Here we interview Ken Behan, Vice President, SYSTRAN Software Inc. and understand what sort of challenges we face when it comes to a multilingual platform like the Internet. We will be asking him about the process involved with translations and analysis apart from the levels of accuracy. Lastly, he will be talking about the company and what purposes it can serve, towards the common good of this society.
Heading into 2016 there has been an increased awareness of the threat of possible data breaches and IT security threats worldwide. Following a challenging 2015 that was characterized by multiple large retailers experiencing massive data breaches as well as federal security threats, it is thought that more than 16,000 cyber attacks will be attempted in 2016. The Federal Government recently released the 2016 Data Threat Report which collated data from IT experts and government entities to discuss and ranks the biggest security threats. The study found that 90 percent of those surveyed believed their organizationwas vulnerable to a data breach. Of that 90 percent, 61 percent experienced a data breach including 1 in 5 within the past year. That is a startlingly high assessment for the experts charged with keeping the nation’s data secure. We took a look at the top security threats exposed early in 2016 to help your company identify arising security challenges, possibly within your own organization. Here is what we found:
1. Increase your deflection rate by translating your self-service knowledge centers into multiple languages. 2. Improve satisfaction with the self-service experience by setting expectations up front. 3. Scale customer service into new regions by supporting additional languages. 4. Respond to … Continue reading →