SimulTrans Localization Blog: SimulTips

Use Machine Translation to Maximize your Budget

[fa icon="calendar"] June 23, 2020 / by Laura Casanellas

Maximize translation budget with MT

With a large percentage of the industry and services on hold, companies’ global budgets might have been frozen or reduced for the foreseeable future. Still, as the world batten down the hatches in wait for a resolution on the global pandemic, companies need to stay current and prepare for the time when economies spring to their feet again.

It is time to use localization budgets wisely, and to find methods of making them stretch as much as possible.

Machine Translation (MT) is a versatile localization instrument that can help:

Depending on the requirements of the project at hand, MT can have more or less prominence on the final text, providing various degrees of savings to the user.

Here are a few examples:

Machine Translation for gisting purposes

Knowing the end goal of each project can lead to define the type of quality that is required, and this will ultimately inform the investment needed in the localization project. For instance, if an affordable localization solution is required for internal communications in a global company, Machine Translation can be used in its raw form, without the need of any human intervention or post-editing.

Most MT providers offer plugins that can be connected with the company’s email application or internal communications system. If such MT engines have been trained with the company’s lingo and terminology, the quality of the outcome will be an improvement if compared with the outcome of a purely generic engine. Using a private account and accessing the MT platform with a unique API key, ensures that the content is translated within a confidential environment.

This use of Machine Translation is called gisting as the aim is to get the gist or general idea of the information communicated. The cost involved here is the cost of creating, hosting and maintaining the engine, as well as the cost of the machine translated words.

Machine Translation for publishing content with a short shelf life

Machine Translation and a light Post-Editing (PE) of the raw output can be used with content that might not have a long shelf life or that needs to be published quickly. For instance, wiki articles can be machine translated and light post-edited if the goal is to publish them in an inexpensive and fast manner. Some companies publish their articles following this method and then do a full post-edit on the most popular pieces at a later date.

The aim of light PE is to ensure that the original meaning is well conveyed in the target translation. This type of post-editing does not pay too much attention to questions of style, in this case, accuracy prevails over fluency. It requires specific guidelines that will be closely related to the end user’s needs.

Machine Translation & Post Editing for publishable quality content

Any other content, with the exception of copywriting or highly creative marketing, can be localized with the aid of Machine Translation. If the goal is to produce publishable quality (near human quality), full post-editing is required. The final outcome of a full post-edited text will be on-par with human translation quality.

Full post-editing is performed by expert post-editors and, once post-edited, the content goes through all quality checks agreed with the client:

  • automatic QA
  • review by a second linguist
  • final formatting checks
  •  any other customized check

Projects with a large number of new words or projects with on-going hand-offs can see real benefits from using machine translation, especially if the MT output is produced by a customized engine.

If you would like to explore if your company's content is suitable for Machine Translation, and what kind of savings you might get, contact us and get a free report.Get Your   Machine Translation   Suitability Report

Topics: Localization Technology

Laura Casanellas

Written by Laura Casanellas

Laura Casanellas is a Machine Translation consultant and specialises in deployment and customization of Machine Translation (MT) programs. Previously she worked in a variety of roles (Language Quality, Vendor Management, Content Management) and verticals (Games, Travel, IT, Automotive, Legal) and acquired extensive experience in all aspects related to Localization. Since 2011 Laura’s focus is on Language Technology and Machine Translation.