SDL Integrates Next Generation Content Translation AutomationLast summer, SDL (news,site) launched a Worldwide Language Service Provider Partner Program to help organizations get their content in the right format to the right customer. This week, they have announced another strategic partnership with Language Weaver, a self-described human communication solutions provider.

SDL plans to deliver Language Weaver’s translation software integrated within their Global Information Management platform and eventually, as part of SDL Knowledge-based Translation System.

Expanding Translation Services

With a growing demand for translation services, SDL wants to evolve their translation management tools and have decided to leverage Language Weaver's expertise by integrating its broader language support with SDL's.

Language Weaver's automated translation solutions have been widely used across the enterprise to deliver translations of digital content. With this partnership, SDL's customers will have the opportunity to deliver effective communication to customers in many more languages, as well as the following functionality:

  • Translation of content that is otherwise too expensive or time consuming to translate
  • Integration of automated translation into the translation process for high-quality technical documentation     
  • Access to automated desktop translation technology called SDL Trados, and integration into

Statistically-Based Translation

Language Weaver's enterprise translation products are built on a statistically-based technology and have been deployed within organizations around the world. Designed to allow organizations to translate large volumes of information into one or more languages at high speed and high accuracy, Language Weaver's products can be integrated with many different applications and interfaces. 

Its translation system learns the statistical relationships between two languages based on the samples that are fed into the learning system.  Because it looks for patterns, the more samples the system sees, the stronger the statistical relationships become.

Translated material can come in any number of formats, and often ranges from translation memory data and glossaries to translated archives and target language texts that are available on intranets, hard-drives and websites.

Translated data is collected, and parallel documents are identified and aligned sentence by sentence to create a “Parallel Corpus.” The Language Weaver Learner processes this corpus and extracts statistical probabilities, patterns and rules, which are called the “Translation Parameters” and “Language Model.”

The Translation Parameters are used to find the most accurate translation, while the Language Model is used to find the most fluent translation.  Both of these components are used to create a new language pair and become part of the delivered translation software for each language pair. 

From Partnerships to Human Translations

From cryptography to algorithms, SDL customers will be aided by Language Weaver's ability to translate automatically from existing translations based upon human translations.