2017 marks one of the major breakthroughs in linguistic technology. This technology is known as Neural Machine Translation or NMT. Unlike the statistical machine translation (SMT), NMT aims to revolutionize the ability to translate languages and this will create a radical impact on businesses today.
Two words: Big Data.
The growth of the internet and connectivity has reached a level where it plays a vital role in the global economy.
The impact on businesses is astounding. From the way new companies are set up, their operational approaches, marketing efforts, globalization strategies, and communication to developing new product ideas and establishing collaborations with other organizations across the globe, this had led to the explosion of big data.
There’s a need for deep learning technology and representation learning technology focused on classification and analysis of raw data.
What makes Neural Machine Translation different?
Neural Machine Translation leverages on neural networks using deep learning technology and representation learning.
What does this mean?
It is no longer about word-for-word translation, that is, the conventional methods of translation, where the target language will seem gibberish to a point that the sentence does not make sense.
It has now shifted to content translation, and this is where NMT comes into effect. It is as good as the machine being able to “understand” the meaning of sentences used. With artificial intelligence, similar to the human brain, NMT will learn from a series of stimuli over a period of time, and run through complex algorithm to be able to produce translated text almost close to human translation.
That is what makes NMT different.
Dale Carnegie once said, ‘to understand someone is to repeat back to them what they said better than they originally described it.’
This applies to all businesses as with individuals.
Does that mean that human translation will be obsolete? Not necessarily. Neural Machine Translation may be powerful and easily accessible, whether in the near future, or even now, but it is not perfect.
There are still challenges that require further research, understanding, and development when it comes to translation, especially for businesses.
Branding: Terminology and Voice
Brand is the image of the company and it is one of the most factors to consider, when dealing with translation.
Brands have their own special terminology and voice. At this moment, NMT is unable to determine communication at this level. Particularly when it comes to marketing messages that may be sensitive in nature, it is still necessarily more effective for companies to engage professional translation services where the human factor is able to distinguish brand values.
Languages to be translated may not always be in a professional manner. NMT is unable to take into account colloquial sentence structure or slang and relate it to specific language and culture.
This may not be relevant for all businesses but for those, for example retail or F&B companies, whose marketing collaterals and advertisements works best when the message is conveyed in their local and native colloquial languages, NMT is not there yet.
From Content Translation to Context Translation
Most likely the next biggest challenge to tackle is the ability for NMT to determine the context of source.
This will cause major concerns for businesses if they were to solely rely on NMT for their business.
Looking at Neural Machine Translation as a whole, business owners need to know when a good time to use NMT for their business is and when is a good time to still rely on human translation.
The benefit of engaging translation services from a professional language solutions company like Transn International Singapore, is the ability to understand the psychology of each translation project requirement. When it comes to translation for business, we provide high quality translation with quick turnaround time.
Transn provides TOTAL language solutions…Connecting The World Through Our Internet Of Languages