"Q&A at the "Machine Translation and Post-Editing Utilization" Seminar"　

On November 17, 2020, we held a web seminar with the theme "Why not review your translation workflow? Utilizing machine translation and post-editing" and had many participants.

At the seminar, we talked about the cost-effectiveness of machine translation, post-editing, and customer case studies.

Here, we will introduce the questions and answers received at the seminar.

Q Post-editors are actually hired, how much translation experience do they have? Do you train post-editors for English to Japanese?

A Years of experience and achievements alone are not enough for us to judge. We test overall language abilities such as grammar level and hire those who meet certain criteria as post-editors. We also use the same hiring and training methods for English-Japanese as we do for Japanese-English.

Q Is post-edit necessary even for translations of GCP (Google Cloud Platform)?

A Rather than focusing on the engine and platform, it is necessary to determine the need for post-editing based on the desired quality requirements and level.

Q Is the post-editing hearing sheet included in the downloadable materials?

A No hearing sheet is included, but we have prepared check sheets and post-editing examples that can be used during post-editing work. Please make use of them.
https://www.science.co.jp/localization/dl/index.html

Q In the case study, it was only up to the second year, but is it possible for costs to scale in the third, fourth, and fifth years as well?
If past translations continue to accumulate, I think the learning effect will increase and costs will scale with each passing year. What do you think?

A As you say, as past translations accumulate and learning effects increase, cost benefits will become even more apparent over time.

Q Is a multilingual engine necessary for implementation?

A It is recommended to determine whether multilingual support is necessary based on the language of your overseas expansion destination.