Machine Learning – A Communication Partnership Moving Forward
In a recent Korea Times article in South Korea, a discussion to answer to question if AI will replace translators has everyone worried in Korea. With English a priority in Korea, a large majority of English majors worry the translation industry will focus on artificial intelligence to do the job especially with advances in machine learning and the Natural Learning Process (NLP). Should English majors be worried if they will have a job available for them when they graduate? Will AI take over the translation industry and take out the human element?
The simple answer is no, machine learning will not replace human translators anytime soon. This is based on a number of reasons.
Language is constantly evolving with new words, new concepts, and even new grammar structure; not to mention incorporating slang and historical word usage properly
Language conversion is always different with humans, so replicating what is the right translation is complex
Computers still cannot fully understand context to properly define word usage within a sentence and with cultural influences, it becomes even more challenging
I think a simple way to put it is that language is not simply ones and zeros like programming. It is more of a gray area instead of black and white. But this is not to say that machine learning is not improving.
The more important question is how can we approach this to create a positive communication partnership moving forward. How does machine learning help translators with translations? Well, it can help with efficiency but more so with the areas outside the actual translation itself. Let’s take a look at the two ways machine learning can help improve the translation industry.
Translation Basics – machine learning is becoming more efficient in providing the basic translation structure, which is helping create efficiencies for the translators as companies find ways to improve productivity in a saturate language market. This approach is perfect for the translator who can then work out the finer details and nuances of the final translation incorporating the human element.
Automatic workflow process – This is where machine learning will play a bigger role in a translator’s efficiency by helping with tasks outside the actual translation role itself. This can include helping write emails, text messages, collecting data, building glossaries and other admin level tasks. Similar to any company, it is these admin task roles that tend to consume a large amount of our time versus working on the actual task at hand.
As we work with artificial intelligence and machine learning, we can see the benefits to make each translation project more efficient and faster, with a focused translator creating more accurate translations and more cost-effective results in the end. Machine learning is helping the industry not only by helping with the translation component but more so by creating efficiencies without even translating a single sentence.
Subscribe to Tinman Kinetics
Get the latest posts delivered right to your inbox