Auto-Translation+

This technical note discusses Lexbe's Auto-Language Translation+ Feature. Please contact the Professional Services department or Sales if interested in Auto-Language Translation+.

Overview

The Lexbe Auto-Translaton+ service uses neural network deep learning techniques to produce more accurate and fluent translation than traditional statistical and rule-based translation models. The neural machine translation system takes into account the entire context of the source sentence as well as the translation it has generated so far, to create more accurate and fluent translations. In comparison, conventional phrased-based machine translation only translates within the context of a few words before and after the translated word. The underlying AWS engines are trained on a wide variety of content across different use cases and domains to perform well on many kinds of content for industry-leading results.

Benefits of neural machine translation:

    • Improves machine translation quality

    • Highly accurate and reliable

    • More natural sounding translations

    • Consistently fast

    • Cost-effective for legal review

Proposed hybrid workflow:

    • Foreign documents segregated

    • Machine translation of all foreign documents to English

    • Review in English (code for Responsive, Privilege, PII, Issues, Key Docs)

    • Produce (in original language)

    • Translate Key Docs by certified human translator

    • Depos

Benefits of hybrid workflow:

    • Reduced review costs

    • Speeds review and production

    • English-speaking attorneys have more control and faster access to key case evidence

    • Key docs are human translated to English, substituted in the document record and can be used in Depos

The Lexbe Auto-Language Translation+ service requires a qualifying Lexbe eDiscovery Platform hosting account. $500 minimum per order. Pricing is per translated page of documents which have been identified using Lexbe’s Auto-Language Detection+ service, or by Customer by custodian or batch. Requires native documents (not images or other documents requiring OCR) for best quality language auto-translation. This service identifies and translates from the following non-English languages:

Arabic, Chinese (Simplified), Chinese (Traditional), Czech, Danish, Dutch, English, Finnish, French, German, Hebrew, Hindi, Indonesian, Italian, Japanese, Korean, Malay, Norwegian, Persian, Polish, Portuguese, Russian, Spanish, Swedish, Turkish

For information on auto-language detection, please see Auto-Language Detection+