Machine translation defined: a computer system that transforms a sequence of characters into another language without human intervention.
Those last three words may give you pause: without human intervention. However, as we’ll discuss, we believe that humans should still be looking over a machine engine’s shoulders to monitor translation quality.
Benefits of using MT include cost savings and speed of translation turnaround times. However, MT may not always be the best solution depending on the type and volume of content you’re looking to get translated and the languages it needs translating into.
There are instances where machine translation without human intervention (also known as raw MT) makes sense for your company. And there are many instances where it does not. Let’s address these issues, along with different types of machine translation engines.
Optimal Content for Machine Translation
We suggest utilizing machine translation for lower-level type content that isn’t business critical. We would not suggest using machine translation for creative, client facing marketing content (without heavy intervention by professional, human translators). The chances for mistakes are simply not worth the cost-savings.
But FAQs, technical documentation and product descriptions, where accuracy is more important than style, can be a good fit for machine translation. A human post editing step after machine translations have been completed, ensure that nuances, accuracy and brand consistency are maintained.
Our Machine Translation Evaluation Process
The need for machine translation is not a simple as offering a list of projects with boxes checked “yes” or “no.” However, the Acclaro team looks at each project and weigh three important requirements before deciding upon machine translation as an option:
- Requirements analysis – Here, we evaluate: content volume, content types, language combinations, quality and privacy needs.
- Content analysis – Is the content ideal for machine translation? To make sure, we check file types, formatting and structure.
- Machine Translation Platform analysis – Depending upon steps 1 and 2, we evaluate which MT platform is ideal for each project.
There are many different types of machine translation engines available. From Rule-Based Machine Translation (where there is signification structure) to Statistical Machine Translation (used by Google and Bing) to Neural Machine Translation (which continuously learns as it translates). We, as your translation agency can help decide which form of MT is right for your project.
This order of MT analysis is imperative. (Why even look at machine translation options if your company doesn’t need it?)
Of course, timelines and budgets are always considered. If you have highly organized, lower-level content that is suitable from a technical and linguistic perspective, then machine translation may work for you to save time and money. It’s simply more efficient than human translation.
If you’re short on time and your budget is tight, then yes, machine translation may work, unless your content is external facing and imperative to your brand’s image (such as advertising campaigns). Then you’ll have to weigh the pros and cons.
Be aware, however, that if your project can utilize machine translation, that there are many types of machine translation platforms. And most all can be “trained” on the terminology, style and voice of your brand to save more time, improve consistency and translation quality.
In our experience, we’ve found that offering a mix of machine translation services with human post editing creates the best results.
To truly see if machine translation is right for your company, you need a translation partner that has not only works with multiple machine translation engines, but is engine-neutral. As we’ve stated earlier, there are many engines out there—some great, some not-so-great—and machine translation requirements can change per project.
Contact an Acclaro professional today to see if machine translation is right for your needs. And if so, we’ll find the ideal one for consistent, efficient translations to help grow your business.