The Anatomy of a Standard Project
Glyph has active projects going in more than 100 languages, but a small subset of those represents most of our work. Here are our core languages, which we work in every day. If you need something you don’t see here, just ask and we’ll source it for you!
Many of our clients ask us about machine translation (MT), whether we use it, and how it is different from using CAT tools with translation memories. While CAT tools use technology to streamline the translation workflow at many stages, the output is ultimately generated by human linguists who can analyze content for diction, syntax, context, tone, and dialect. Machine translation instead uses computers to analyze and translate text automatically.
Glyph does not currently employ machine translation in production. We have been investigating and researching the possibility since 2013, and in 2015 we designed and ran a study in conjunction with the National Institute of Standards and Technology (NIST), providing quality assessments on machine translation output using human linguists as judges. We have found that at the current state of the art, MT output still requires significant human editing to produce the level of quality that we consider requisite for the kind of projects we undertake.
Machine translation technology is advancing quickly, and for high-volume, low-impact applications such as eCommerce platforms with millions of products, current MT capabilities already make sense. There are language service providers who use machine translation effectively in the right context. And that’s really the crux of the MT question – it all depends on the context. There are already applications for which MT makes good business sense, and there will always be some creative applications for which human linguists will be the gold standard. For the moment at least, Glyph remains committed to professional human linguists.
LET’S DO THIS!
Don’t need a quote, just want to talk?