While you’re never too old to get to grips with a new language, becoming fluent will be a tall order if you start as an adult, even if you immerse yourself. The brain’s neuroplasticity - referring to its ability to adapt and create new neural pathways - decreases with age, giving adult language learners a considerable disadvantage over children.
Learning a language from a young age is imperative if fluency’s to be achieved. Many European countries - including France and Spain - start English classes as part of their core curriculum as soon as children start school, meaning that they can converse in the language with a degree of fluency from a young age.
In order to aid language learning, a number of translating services are available online. Most - like Google’s standard translator tool - require the user to type in the word they’d like to translate. Often, the results are spurious, with the software unable to translate full sentences into natural language, instead doing it word for word. Yet that service is Google’s most basic iteration. Last month, the company announced that its translate app had incorporated real-time visual translation compatible with more than 20 different languages.
Using deep neural networks - the more powerful form of regular neural networks - the application can read road signs, menus and billboards and translate them into the user’s native tongue. Considering that computers couldn’t recognize the difference between a cat and a dog five years ago, it really puts into perspective how far recognition technology has come.
Regardless of how powerful these applications are, when they’re most needed fate often intervenes and decides that your internet connection is going to falter. Thankfully, the translate app doesn’t require an internet connection - a big perk considering that people would presumably be unwilling to pay for roaming charges when they go on holiday - and uses deep learning to distinguish between a letter and a background object.
What proved to be one of Google’s biggest challenges was getting the app to recognize letters which weren’t clear. The number 5, for example, could be easily mistaken for an ’S’, but even if the app were to read the word ‘5oda’ it would be able to tell that the word ’soda’ was what the user had initially wanted to translate.
Whether a service such as this can have much of an impact of language learning remains to be seen. Having a readily available translator service arguably makes people less inclined to memorize words, especially if the service proves reliable. This sentiment was echoed by o Julie Cattiau - Google Translate’s Product Manager - who said;
‘Translate is not a replacement for a language-learning course; it’s not going to teach you a language from scratch or all of the intricacies involved in learning a language.’
It can however be used as an assistant and a way of helping language learners discover new vocabulary whenever they see something they don’t understand. It’s clearly an upgrade, however, from the standard translator service and could make ordering that meal in France this summer just the little easier.