AI & Machine Learning

New Translate API capabilities can help localization experts and global enterprises

google translate api.png

Whether they’re providing global content for e-commerce, news, or video streaming, many businesses need to share information across many languages, and increasingly they’re turning to automated machine translation to do it faster and more cost-effectively.

As one of our longest-standing AI products, Google Cloud Translation has substantially evolved over the years to meet the increasing needs of developers and localization providers around the world. Last year we launched AutoML Translation to help businesses with limited machine learning expertise to build high-quality, production-ready custom translation models without writing a single line of code.

But that only met part of the need for customization. While AutoML Translation offers full customization flexibility, companies using Translation API wanted to have more granular control on specific words and phrases, such as a list of location names or street names. Many businesses also told us they use both custom and pre-trained models in the same translation project for different languages, so we wanted to make it easier for them to move between models to run translation predictions.

As a result of these learnings, we recently launched Translation API v3 to better serve the needs of our customers. Here’s a look at the enhanced features in Translation API v3.

Define specific names and terms with a glossary
If you need to maintain and control specific terms such as brand names in translated content, creating a glossary can help. Simply define your company-specific names and vocabulary in your source and target languages, then save the glossary file to your translation project. Those words and phrases will be included in your copy when you apply the glossary in your translation request.

Welocalize, a global leader in multilingual content and data transformation solutions, is already using the glossary feature to increase accuracy, efficiency and fluency for their customers. “The new Google glossary feature will have a significant impact on our day-to-day business. We process hundreds of million words per year using machine translation in widely disparate enterprise client scenarios,” said Olga Beregovaya, their Vice President of Language Services. “The ease of customization and API consumption allows us to enforce broad terminology coverage for both clients with voluminous data in Google AutoML Translation and clients with sparse data in Google Cloud Translation API. Our initial benchmarking in five languages shows a preference for translation with glossary as much as 20% over the non-glossary.”              

Select between custom and pre-trained models
Now you can choose between Translation API’s traditional pre-trained models or use custom model translations so that you can streamline your workflow within the same client library.

Streamline your localization process with batch translations
You can now translate larger volumes of content in one translation request for text and HTML files stored on Google Cloud. This means you can use a single request to upload multiple files translated into multiple languages using multiple models.

For example, if you wanted to translate an English product description on your website into Spanish, Japanese, and Russian, you could use your custom AutoML model for Spanish and a pre-trained model for Japanese and Russian. You would simply upload your English HTML file to your Cloud Storage bucket and send a batch request pointing to your Spanish AutoML model and pre-trained models for Japanese and Russian. Translation v3 will then output your HTML to Cloud Storage in Spanish, Russian, and Japanese in three separate files.

Integrations with TMS/CAT
Technology partners are starting to integrate these new features within their TMS systems as well.

“Based on feedback from our clients using Google translation technology in CAT tools, the most sought after features are the ability to customize Google Cloud Translation output with glossary and to make translation faster via batch translation,” says Konstantin Savenkov, CEO of Intento. “Now both are available via our plugin to CAT Tools (SDL APPStore, MemoQ, and Matecat). Also, we may deliver Translation API v3 to enterprise TMS systems via our XLIFF-based connectors." You can learn more in Intento’s recent blog post.

Introducing the Translation API v3 free tier
Starting with Translation API v3, businesses can take advantage of a free tier for the first 500,000 characters per month. Beyond that, the pricing remains on a simple and affordable per-character basis, so you only pay for what you use. You can learn more on our pricing page.

How to get started
If you’re already using Translation API v2, you can begin migrating your applications to v3 using this guide. For more information on Translation API v3, visit our website.