Elodie Vermant, a Swansea University student, studying for an MA in Professional Translation, shares her experience on using Machine Translation for the first time at Swansea University.

The MLTM11 Translation Technologies Module is taught by Dr. Maria Fernandez-Parra, Lecturer, Languages, Translation and Communication at Swansea University. Read more experiences from her students.

KantanMT: What made you choose this course, and how to do you think it will help your career?

Swansea_UniversityI chose this course for various reasons. First, for the location: it is based in the UK, and I thought that it would be more interesting for my career to graduate in an English-speaking country, as I am studying English-French translation. Swansea University was also cheaper than most UK based universities, and I really liked the fact that it is located between the beach and a park!

Secondly, this course has a good reputation and is affiliated with different interesting programmes such as the METS programme. I also appreciated the fact that they accepted students with only one language combination, and the possibility to do an internship instead of a dissertation: less theory and more practice, good for employability!

KMT: Had you any previous experience of Machine Translation, if not how did you find using the KantanMT system?

Elodie Vermant: I am personally not a big fan of Machine Translation, so I can say that I do not have any previous experience with it. I simply think that it is faster and more efficient to translate something from scratch than using any MT tool and post-edit. KantanMT is interesting, but I would still not use it to translate.

KMT Responds: Many experienced translators have similar impressions when post-editing Machine Translation output for the first time.

The reason might be related to the fact that translating and post-editing are two different cognitive tasks; this is often why some translators perceive that post-editing takes longer than it really does.

The reality is that, in general terms, and at least for technical related content (Technical documentation, UI, knowledge bases, product descriptions, etc.) post-editing is faster than translating. We don’t include more creative content like copywriting, as in this case, content sometimes needs to be almost re-written or transcreated, rather than purely translated.

KMT: Can you describe your assignment and thoughts about using KantanMT.com, what did you find good and do you have any suggestions for improvement?

Startup Stock Photos

EV: I like the fact that the project management is automated, and that I can receive emails every time my translation is ready. I also like the quality leverage included such as the BLEU automatic metric. My only concern is regarding the translation job itself. I tried to translate a small text, and the result was not good enough: a lot of modifications were still necessary.

KMT Responds: KantanMT has many functionalities that make it highly customisable. It is advisable that feedback from post-editing is gathered and used to improve the engine by using some of the post-processing or terminology related functionalities. For more information, please visit these features in our KantanMT Help menu. A series of webinars focusing on maximising KantanMT features will kick off in August.

KMT: What is your impression of the translation industry and in your opinion, what do you think the industry will look like in the future?

EV: I think that the translation industry is promising, and more and more translator jobs will be needed in the future. I wish it could be more structured and regulated, and also more recognised, as I still find many people telling me “why are you doing an MA in Translation, while people can just use Google Translate?”. There could be a place for MT in the translation industry in the future, but a lot of improvements are necessary, in my own opinion.

KMT Responds: MT has already achieved a prominent space in the translation industry, in some cases it is used in conjunction with professional post-editing, in some others it is used to produce output that is published directly without any further human intervention. Machine Translation can be considered as a productivity tool that supports translating large amounts of information produced.

KantanMT is used by Swansea University as one of the language tools used in its translator training courses: MA in Professional Translation and an MA in Translation and Interpreting.

To find out more, or join KantanMT’s growing academic partner programme, contact Louise Irwin, louisei@kantanmt.com.

Download Partner Brochure (PDF)

About Swansea University

Swansea is a top-30 UK research university (Research Excellence Framework 2014), with a record of teaching translation tools and technologies going back to 1999. Its MA in Professional Translation (formerly MA in Translation with Language Technology) has been a member of the European Master’s in Translation Network since its inception in 2009. It also offers a vocational MA in Translation and Interpreting, and a thriving PhD programme in Translation (including recent successful projects on translation tools).