SYSTRAN’s exclusive customization methodology combines the latest linguistic and statistical techniques that are used to train the software to customer domains and improve translation quality. This customization process includes the creation of dictionaries and the building of translation models from existing data. The software learns how words and phrases should be translated for specific domains based on existing human translations. Because SYSTRAN’s hybrid MT solutions combine rule-based and statistical techniques, the amount of data required to train the software is much less than pure statistical MT solutions.
Benefits of Customization
Customization is a process that allows users to fine-tune translations to achieve publishable translation quality results.
Customization is appropriate for companies with high volume translation needs in a specific domain, such as travel, automotive, IT, industrial automation, patent research, and consumer electronics. Any industry with specific terminology and high volume or continuous translation needs will benefit from customization.
Customization Process
The SYSTRAN customization process leverages existing corporate language assets to create domain-specific dictionaries and translation models for each language pair. The corporate assets leveraged by the customization process are:
- Monolingual texts
- Bilingual texts
- 翻訳メモリ
- Corporate terminology
Customization deliverables consist of a set of translation resources that are used during the translation process. Use of these resources generates publishable quality translations in the desired domain. This allows companies to translate data that has never been translated before and deliver quality multilingual content.
For more information about the customization process or to discuss your company’s translation needs please contact us.
