On April 18, we will present "What Translation Companies Can Do to Alleviate the Stress Felt During NMT Post-Editing" at TAUS.

April 10, 2019

On April 18, we will present "What Translation Companies Can Do to Alleviate the Stress Felt During NMT Post-Editing" at TAUS.

On April 18, 2018 (Thursday), we will present "What Translation Companies Can Do to Alleviate the Stress Experienced During Post-Editing of NMT" at the TAUS Global Content Summit Tokyo.


The translation accuracy of Neural Machine Translation (NMT) has significantly improved compared to previous rule-based (RBMT) and statistical-based (SMT) machine translation. This technological innovation has led to an increasing level of customer requirements for localization, both in terms of quality and delivery time.


On the other hand, post-editors who correct machine translation outputs are feeling stressed. One of the issues with NMT is the "inconsistency of terminology." Therefore, post-editors are taking great care to use the correct terms. However, with the emergence of customizable MT engines recently, the terminology issues may be resolved in the near future.


So, will the errors in the style guide also be resolved with the customizable MT engine?


In Japanese translation, the style guide has many more rules compared to other languages. Furthermore, the style guide varies depending on the client and the product. As a result, the style errors and complex style rules included in NMT translations can be a significant source of stress for Japanese post-editors.


The application of terminology in NMT has been discussed in various contexts, but the issue of style has not been addressed much until now.


This presentation will explain how human science has addressed this significant challenge and supported the improvement of post-editor productivity.


taus-logo-news






TAUS stands for Translation Automation Users Society, and it is a membership organization composed of members from companies related to translation. It promotes innovations such as machine translation and cloud translation.

 

 

Event Guidelines

Date and Time
Location
  • Oracle Aoyama Center
  • 2-5-8 Kita-Aoyama, Minato-ku, Tokyo
Costs and Participation Methods

Speaker

IMG_2787_2

Toru Shishido

Involved in translation work at a document production company since 2006.

After working at a foreign-affiliated translation company, I joined Human Science in May 2017.

Has served in roles such as project manager, translator, and reviewer.

Additionally, we have experience in supporting DITA implementation and conducting initial quality assessments of machine translation engines.

Engaged as a solution consultant at Human Science.

 

 

Machine Translation Achievements

WingArc1st Inc.

NIDEK CO., LTD.

Cross Language Inc.

Domestic electronics manufacturer (listed on the Tokyo Stock Exchange First Section)

Foreign IT Manufacturer

Related Services, Blog

List of Machine Translation Solutions

Machine Translation Integration Solution MTrans for Trados / MTrans for Memsource

Cloud-based AI Automatic Translation Software MTrans Team

Automated Post-Editing Tool MTrans Post-Edit Booster

Language Technology Staff Blog

Open Seminar Scheduled

Open Seminar List

 

Contact Us

If you have any questions regarding the content of this presentation, please contact hsweb_inquiry@science.co.jp .

For inquiries by phone, please contact us at the number below.

TEL: +81 35-321-3111

Person in Charge: Business Promotion Department, Saga