Join the DZone community and get the full member experience.Join For Free
The ability for the crowd to perform a wide range of tasks is pretty well established, with a huge number of case studies from different industries to choose from. One that has proven particularly effective has been that of translation. Duolingo was undoubtedly the forefather of the crowdsourced translation movement. Launched in 2011, the site combined the captcha spam traps that are a common feature on the web, with that of translation. So basically, when we fill in captcha forms, we’re helping to translate text, which is then used to help people learn a particular language.
With the rise of smartphones over the past few years, there have been an increasing number of apps that look to help us out with our language woes whilst on the move – in a foreign country for instance. The latest of these is the Translate.com app. The app aims to help users make sense of any piece of language, whether written or spoken, and aims to do so via the power of crowdsourcing.
The app, which is currently available on both iPhone and Android devices, allows you input text in a number of ways. You can either type or copy written text, record some speech, or even upload a photo of some text. The app will then detect the language being used, and offer up a translation of the text into one of 75 other languages. This initial translation is performed by an algorithm, but if you’re not satisfied with what the computer gives you, you can turn to one of the 2 million members of Translate.com for a human translation. The text can be translated by multiple members, with each attempt then rated for accuracy by fellow members.
Whilst machine translation has come on leaps and bounds in recent years, this human element will retain importance I think, at least in the short-term. This is especially the case for languages, many of which have subtle nuances that computers find hard to detect, yet which humans indelibly learn.
The next challenge will be to speed up the process and attempt to provide real-time, human translation, perhaps via a Mechanical Turk style network of humans around the world.Original post
Opinions expressed by DZone contributors are their own.