Can Machine Translation be used to translate all content?

You may consider using MT for all content, but its use ultimately depends on the type of content and its purpose. 

MT has a place in every workflow as an efficiency tool to decrease translation costs, increase translation speed, and boost content velocity. However, companies will achieve their content goals best by blending MT with the services of human translators. 

MT — with little-to-no human intervention — is particularly suitable for lower-visibility content that doesn’t need perfect translations. MT can support translators handling large volumes of content that require higher quality levels.

What is the difference between Machine Translation and Neural Machine Translation?

Machine Translation is the automatic translation of text from a source language into a target language. Neural Machine Translation (NMT) is a translation technology that uses neural networks when it automatically translates content from a source language into a target language. 

NMT results in a higher level of accuracy because it translates material within a broader context as opposed to only addressing individual words.

When should I use Machine Translation?

In certain situations, engines are superior to human translators. MT is ideal when you must translate high volumes of low-visibility content fast.

A good example is the translation of user-generated content, such as reviews. MT is also helpful when translating manuals and documentation, some knowledge base articles, or simple landing pages. 

As MT technology further evolves and improves, it will be able to handle more sophisticated content and use cases.

Is it risky to use Machine Translation? 

While MT quality has improved over the years, engines are still prone to making errors. The more valuable and visible your content is, the riskier it is to use machines for translation without human intervention. We advise our customers to incorporate post-editing by professional human translators for high-value, high-visibility content to prevent errors that can lead to reputational, financial, or legal harm. Read our blog, Machine Translation and Catastrophic Errors, to learn how to avoid catastrophic MT errors.

Is Machine Translation secure?

Free MT offerings like Google Translate or Bing Translate are not deemed safe for sensitive data since the solution may use your material to train its engines. Your data is the price you pay to use these offerings. 

If you are working with internal business, legal, or private documents, consider using paid solutions that provide more privacy and security. 

Will Google be satisfied with Machine Translations that are refined only slightly by human translators during post-editing? 

This question was posed to Google during an online hangout by a company that uses MT to provide posts in other languages. As reported by Search Engine Journal, Google’s Lizzi Sassman answered as follows:

“Well, that’s good to hear that the human translators are [primarily] happy [with the Machine Translations], and this is [minor edits are] totally fine for Google as long as there’s a human involved in the review process. That’s the key. The thing you want to watch out for is making sure that the quality continues to be good and working well for the humans that are reading the content.”

Companies must produce content Google will view favorably. Google’s response underscores the value of MT when you monitor the quality of the output to ensure readers will like it.

How can I start using Machine Translation?

As Machine Translation experts, Lionbridge can help you start using MT to get the most out of the technology without incurring detrimental errors. Among our services, we can help you determine which engine is best for your content, improve output quality with our Smaⁱrt MT offering, provide MT training when required, and put an infrastructure in place to meet varying quality levels.