Machine Translation Quality Estimation

Leverage MT the same way you would leverage TM matches. Using MTQE, quality scores are automatically calculated before any post-editing is done, removing the guesswork from MT and improving post-editing efficiency.

  • Eliminate MT guesswork

  • Faster translations

  • Assess MT engine quality

  • Forecast post-editing efforts

The feature can be used with over 70 language pairs and is available for all MT engines supported in Memsource and in all Memsource editions.

Initial testing indicates that for the supported language pairs, MTQE can provide quality scores of between 85% and 100% in up to 14% of segments that are machine translated. This could mean savings of up to 7% on post-editing costs.

  • The quality of machine translation continues to improve and linguists are increasingly being asked to edit, or post-edit, machine translation output. But the quality of MT still varies considerably, depending on various factors, including content type and language pairs. Linguists have to sift their way through a lot of low-quality machine translation, occasionally

  • coming across something that only requires a bit of editing or none at all. This wastes a lot of time.

    We know that linguists are saving time with quality scores for translation memory matches, so why not provide the same for machine translation. This is where our AI team came in.

Gamehouse - logo
A Return on Investment
“With improved terminology consistency and translation quality, Memsource essentially pays for itself.”
Tommy Nordkvist - photo
Tommy Nordkvist

Head of Game Operations

The MTQE quality scores are available in the Memsource Web Editor and Desktop Editor. In MTQE Version 1, there are four scoring categories:

100% MT
Excellent machine translation quality, probably no need for post-editing

95% MT
Very good quality machine translation, possibly minor post-editing required

85% MT
Good match, but likely to require some post-editing

No score
When there is no score, this means MTQE cannot identify the quality (it may be high or low), so the output needs to be checked by a linguist