Many translators fear that their jobs will become obsolete due to the AI and machines replacing them. Is it true? Or are they just unfounded fears brought by new technologies?
Today, we will discuss machine translation trends, its history, and why translators should look forward to its potential in the translation process.
Now, let's dive right into the heart of the issue and discuss all about machine translation trends.
To start our discussion on translation technology and the recent development of artificial intelligence and translation, let's define what it is and its history.
Machine translation is a specialized technology that translates text from its original language to a target language without human assistance.
The translation technology that translators often utilize is machine translation post-editing, which operates on machine learning and neural networks. It's predicted that we'll see many new players that will drastically affect the translation industry in the coming years.
We have come a long way since a 1950s experiment where a machine translated texts from Russian to English. Medium reported that artificial intelligence and translation technology were developed side by side and have contributed to each other's progress.
Even at the beginning of translation technology, it incorporated neural networks and AI machines to mimic human intelligence. The United States, Russia, and Japan were some of the few countries that invested resources in researching and developing translation software.
Although the media initially sensationalized artificial intelligence technology and its potential, there were several delays in its developments due to a lack of support and funding from the private and public sectors.
But even with these obstacles, it led to the development of third-generation translation technology that became crucial in creating translation technology that we use daily, such as Google Translate.
Recently, with the development of neural translation technology and the advancement of AI in the past decade, many translators have become concerned if their work will be one of the many jobs taken over by AI and robots.
Fortune reported that many experts believe that 40% of the global workforce will be automated and replaced by AI.
Meanwhile, the World Economic Forum has stated that an estimated 30% of companies worldwide will need to retrain their employees for the new technology in place.
Now, it might seem like a bleak future for translators, but it's not.
According to Lionbridge, neural translation technology improves at an average of 3% to 7% per year. And so, we still have a long way before translation technology replaces human translation expertise.
There are certain aspects of the translation process that rely on the knowledge of both the original language and the target language's culture, which we haven't processed with the current technology yet.
This is made clear with BBC's report, which stated that despite machines having the capacity to take in larger quantities of information than the average human, they were still far from perfect.
Researchers discovered that when neural translations were to summarize the results of their search engine results for their user, "they hallucinate." Hallucination for neural network models is when they insert data that is not part of the source.
Even though translation technology is necessary for immediate communication, it can't truly replace the work of translators, especially when it comes to translating vital documents.
In other words, as mentioned by the World Economic Forum, translation providers will be those that will need to retrain their translators if they wish to implement advanced translation technology into their language solutions.
Modernizing and implementing technology in any organization needs a holistic approach. Translation providers must create strategies to train and educate their translators, incorporate the technology in their language services, and routinely conduct maintenance.
National Geographic reported that every 14 days, one language dies and that it is estimated that in a century from now, over 7,000 languages will disappear. This is distressing news because languages preserve cultural identity, history, "unique modes of thinking," and community customs.
Many critics have blamed globalization and technology for being the cause of these languages dying out because the vast majority of global content is either in English and Chinese, which has made it harder for other languages to be adopted in the digital space.
However, the same technology being criticized for the extinction of languages could potentially be the one to save them. Again, translation technology and machine learning can record and take in large volumes of linguistic data, which is why multinational companies like Google and Microsoft have invested heavily in making neural translation platforms and tools.
The adoption of AI will supercharge the work of linguists and anthropologists for the coming years. Besides recording extinct and endangered languages, AI has the potential to teach the younger generations of their indigenous languages.
ARC Centre of Excellence for the Dynamics of Language (CoEDL), an organization dedicated to preserving endangered languages in Australia, created an AI and robot that teaches children indigenous languages through games, class lessons, and stories. The platform monitors and documents the child's progress in learning the language.
However, despite several AI initiatives that have emerged to protect languages, translation technology, and machine learning and machine translation trends and technologies cannot fully capture the essence of a language.
Pronunciations, facial expressions, and dictions cannot be captured even with the most advanced technologies. Not only that but the meanings of the language that require cultural understanding will most likely get lost in translation.
But this is expected once a language dies due to no longer having any native speakers. Like with Latin, we may understand written documents but specific characteristics such as how it's pronounced get lost over time or evolve into another form.
You're able to communicate with people from all over the world, regardless of their language, because of the collective efforts of linguists, scientists, and programmers who sought to overcome language barriers.
In the next industrial revolution brought by artificial intelligence technology, translators will benefit significantly from implementing translation technology in their translation process.
Meanwhile, it's amusing to think that hundreds of years from now, our current machine translation and machine translation trends will become the Rosetta stone of future generations.