SYSTRAN has been wholeheartedly involved in open source development over the past few years via the OpenNMT initiative,whose goal is to build a ready-to-use, fully inclusive, industry and research ready development framework for Neural Machine Translation (NMT). OpenNMT guarantees state-of-the-art systems to be integrated into SYSTRAN products and motivates us to continuously innovate.
In 2017, we published OpenNMT-tf, an open source toolkit for neural machine translation. This project is integrated into SYSTRAN’s model training architecture and plays a key role in the production of the 2nd generation of NMT engines.
Since the publication of the Executive Order on Maintaining American Leadership in Artificial Intelligence by the White House this past February, many government agencies are struggling with getting started in AI. They realize use of this technology will help them be more efficient. However, finding those tasks that will be “quick wins” in moving towards AI adoption is the main challenge.
Last month, we conducted a webinar “So, You Think Your Game Is Localized?”, the first of a 3-part-series given by Elizabeth Senouci from XTM International, and Victor Ramirez from SYSTRAN.
If you couldn’t guess by the title, “So, You Think Your Game Is Localized?” was a webinar focused on Video Game Localization. Senouci and Ramirez are both experts on the topic and thus decided to share their knowledge with the video games community.
In the webinar, Senouci and Ramirez discussed the need for game localization, some basic terminologies associated with it, user interfaces, global marketing, and the importance of customer service.
“Localization isn’t just one thing you can do and just get done with it. It’s a holistic process and it’s actually customized based on your game, your product,” Elizabeth said in her intro.
For staff of multinational companies who want to translate a simple phrase or word, systems like Google or Microsoft come in just handy. They help you order a taxi in Japan, pay your restaurant bill in France, and impress your clients with a hearty “jó reggelt” (“good morning”) in Budapest. The problem is such tools are notorious for imprecise translations and data leaks.
Would you really want to use Google Translate for that internal email to your affiliates in another country?
On the other hand, research from the European Parliament shows that on average a common language increases trade flows by 44%. So, how do you – and your staff – hack through language barriers and achieve professional communication in the business world?
Machine Translation users care about quality and performance. Based on our own observations and the feedback we’ve received; the quality of our Neural MT is impressive. Evaluating performance is a stickier subject, but we’d like to dig our hands in and present our innovations and achievements and how it benefits NMT users.
By performance we mostly mean the manner in which a system performs in terms of speed and efficiency in varying production environments. It is important to note that performance and quality in Neural MT are tightly connected: it is easy to accelerate a given model compromising on the quality. Therefore, when evaluating performance improvement, we always check that quality remains very close to optimal quality.
Since switching to NMT at the end of 2016, we’ve invested our R&D efforts into optimizing our engines to be more efficient, while maintaining and even improving translation accuracy. Our latest, 2nd generation NMT engines, available in our latest release of SYSTRAN Pure Neural® Server, implements several technical optimizations that make the translation faster and more efficient.
New model architecture
The first generation of neural translation engines was based on recurrent neural networks (RNN). This architecture requires the source text to be encoded sequentially, word by word, before generating the translation.
Data leakage and lack of information are two critical issues that can harm businesses. Nonetheless, due to the ever-growing global marketing and communication needs, the temptation to use the fast and free online translation tools are rising.
Apart from the apparent dangers that these tools pose to businesses such as miscommunication, loss of business, and cultural insults, there is critical important threat that many enterprises often fail to recognize.
Whenever an employee uses a free online translation tool, they may cause massive data privacy breaches by making the consumer data searchable. Data breaches as such mainly happen due to employee negligence looking for quick machine translation, and it can often put millions of customers’ sensitive data at exposed on the internet.
Companies thus struggle to find the right balance between enabling business and securing information. Without the capability of translating software, potentially hundreds, if not thousands, of employees could turn to free translation tools to get their content translated in turn making the content available online.
Until recently, using machine translation (MT) was considered a hindrance by serious translators. Now that machine translation is powered by artificial intelligence, translators in the government are intrigued by this new technology. Forward-thinking linguist programs recognize the value of MT, and it’s only a matter of time when others will follow suit. Consider these four reasons as motivation for modernizing the status quo:
1. Translate Smarter
As with many other skilled professions – accountants, doctors, analysts – technology is a time-saver. Translators now have the same benefit. In fact, commercial benchmarks show that neural MT helps translators post-edit at 2000 words per hour. Without technology, which is typically the case in the government, translators translate at 300 words per hour. Imagine the time-savings — the same 6000-word document can now be translated in 3 hours instead of 20. Additionally, SYSTRAN MT will retain the formatting of the original document, which further saves time.
Experience unprecedented integration of customer terminology with neural networks!
SYSTRAN Pure Neural® Server, our state-of-the-art translation technology tailored for businesses, delivers quality, fast, and secure translations using Neural Networks and Artificial Intelligence. We have just added support for a unique feature that takes it a step further. Users can now add custom terminology to be used in their translation tasks. Seasoned users know about User Dictionaries in our previous rule-based and hybrid technology, but this feature was not fully implemented by the Neural Networks. Until now.
Translation tailored to your need
User Dictionaries (UDs) are key in customizing translation to users’ needs by allowing them to determine their own terminology and ensure that it is translated as such regardless of context. They can also be used to disambiguate between a word with multiple meanings. In this case, translation profiles can be created that apply user dictionaries with the ambiguous term translated differently in each. For example, “mettre sous tension” would be translated from French as “to turn on” in a Generic profile, but a user could create an Aeronautical profile and add the entry to a UD as “to energize” and if needed create an Electronics profile for the term to be translated as “to apply power.” User Dictionaries can also quickly correct any translations that are not accurate for the user’s context. User dictionaries are primarily used so that industry jargon and brand, model and product names are translated accurately.
In today’s accelerated globalization, booming e-commerce and customer service digitalization, the languages spoken by potential consumers come in hundreds. Global companies are faced with a problematic equation: while they might have centralized their customer service operations, it is still costly to recruit an agent for every language covered. It is nevertheless crucial to respond to customers in their native language quickly, efficiently and at minimal cost to achieve excellent Customer Service.
Breaking the language barrier
It’s simple: translating client emails instantly, responding to them in their language just as quickly or automatically displaying the most common answers in FAQ databases in multiple languages is a real Game Changer. For global Customer Service teams, the response time in foreign languages can be divided by 10 after Neural Machine Translation implementation! Calls are reduced with increased usage of a multilingual online knowledge base and customer satisfaction is higher than ever!
Data leakage is the biggest threat
More importantly, unsecured online translation tools, often used by customer service teams to understand customer queries in foreign languages expose companies to data leaks. The nature of interactions during customer service operations can be as critical as sharing account information, credit card numbers, passwords and so on…
To guarantee complete security of your customer data, it’s absolutely key to rely on a provider that is able to provide the translation service on-premise or accessible via a private cloud. It is also true for internal support interactions. SYSTRAN also provide translation solutions for enterprise IT departments following a trend of global support services centralization, and allow them to manage technical support requests worldwide while ensuring user data security.
Since 2016, there has been a sharp increase in open source machine translation projects based on neural networks or Neural Machine Translation (NMT) led by companies such as Google, Facebook and SYSTRAN. Why have machine translation and NMT-related innovations become the new Holy Grail for tech companies? And does the future of these companies rely on machine translation?
Never before has a technological field undergone so much disruption in such a short time. Invented in the 1960s, machine translation was first based on grammatical and syntactical rules until 2007. Statistical modelling (known as statistical translation or SMT), which matured particularly due to the abundance of data, then took over. Although statistical translation was introduced by IBM in the 1990s, it took 15 years for the technology to reach mass adoption. Neural Machine Translation on the other hand, only took two years to be widely adopted by the industry after being introduced by academia in 2014, showing the acceleration of innovation in this field. Machine translation is currently experiencing a golden age of technology.
From Big Data to Good Data
Not only have these successive waves of technology differed in their pace of development and adoption, but their key strengths or “core values” have also changed. In rule-based translation, value was brought by code and accumulated linguistic resources. For statistical models, the amount of data was paramount. The more data you had, the better the quality of your translation and your evaluation via the BLEU score (Bilingual Evaluation Understudy, the most widely used algorithm measuring machine translation quality). Now, the move to Machine translation based on neural networks and Deep Learning is well underway and has brought about major changes. The engines are trained to learn language as a child does, progressing step by step. The challenge is not only to process exponential data (Big Data) but more importantly to feed the engines the most qualitative data possible. Hence the interest in “Good data.”