[Case Study] Achieved 37% cost reduction and 50% shorter translation period through optimal engine selection - Office Equipment User Manual

This time, we would like to introduce a case where our company provided consulting for the selection of a machine translation engine.

In 2014, from the person in charge of the instruction manuals at a domestic office equipment manufacturer

・We want to achieve simultaneous release of manuals in Japanese, English, and multiple languages.

・I want to respond promptly to frequent updates of the user manual.
To achieve this, I want to shorten the time required for translation.

For this reason, we received an inquiry about introducing machine translation.

 

——————————————————————————————–

●Project Overview

Translation Target Item: User manuals for office equipment
Language: English → French, Italian, German, Spanish, Brazilian Portuguese
Volume: 400,000 to 500,000 words per year
Corpus usable for statistical-based engines: Approximately 20,000 sentences

——————————————————————————————–

When introducing machine translation, customers should

  It is unclear whether introducing machine translation can truly shorten translation time and reduce costs.

  I don't know which engine to choose from the dozens of commercially available engines.

You had concerns like these.

 

Therefore, our company has proposed three engines that align with the project and customer budget,
and we compared each engine based on factors such as quality, return on investment, compatibility with tools previously used,
and whether the accumulated past translation assets (translation memory data)
can be utilized, and submitted a report.

 

Machine Translation Engine Evaluation Report

 

As a result of the verification, when using the engine with the best quality,
the translation cost can be reduced by approximately37%, and the translation period can be shortened by approximately50%.

This is the effect of cost reduction and shortening of translation periods while maintaining the same level as "human translation"
in terms of quality.

When using the second best quality engine,
the estimated reduction rate in translation costs is about25%.

 

In this way, the reduction rate of translation costs can vary significantly depending on the machine translation engine selected.

The better the quality immediately after machine translation, the less burden there is on human editing (post-editing), which can lead to greater cost savings and reduced translation time.
Depending on the target content and the corpus used,
the optimal engine may vary,
so we recommend conducting quality comparisons using actual data with several engines in advance.

Human Science provides consulting for the selection of machine translation engines from a neutral position without its own proprietary engine.
If you can provide the relevant content or corpus,
we can also conduct quality assessments and cost simulations based on the current situation,
so please feel free to contact us.

 

>> Contact Form

If the form is not available, please contact us at hsweb_inquiry@science.co.jp with your inquiry.

Please send it via email.

Alternatively, please feel free to contact us by phone at TEL: 03-5321-3111.