Do-it-yourself MT: Taking (statistical) machine translation to the next level

New Web-based statistical machine translation services are currently revolutionizing the market. These SaaS solutions allow users to customize an MT engine with their own translation memories. As most of these services follow the subscription model, launching a DIY MT project costs only a fraction of traditional machine translation tools, which makes this powerful technology affordable for even the smallest organization.

Text by Uwe Muegge

Inhaltsübersicht

Do-it-yourself MT: Taking (statistical) machine translation to the next level

Image: © mrPliskin/ istockphoto.com

The statistical approach to machine translation, where large sets of training data determine the translation result, has become the dominant force in automated translation. With Google Translate, which currently supports 71 languages, statistical machine translation (SMT) has become easily available to anyone who wants to use it. For professional applications, tailoring the output of an SMT system to the needs of a specific audience used to be a major challenge – at least until very recently.

During the last two years, several SMT services have emerged that give users a high degree of control over the quality of machine-generated translations. Their business model is simple: Make a powerful statistical machine translation system accessible via the Web and allow users to customize an MT engine by uploading their own translation memories. And reduce the price of using a customized SMT ...