What's new in this release
An overview of the new features and functionality of this ETS product release.
Version 7.4.3
ETS version 7.4.3 fixes some known issues.
- Newlines in the Quick Translate UI are now treated as sentence ending.
- Allow the removal of the last login group in the LDAP Configuration.
- Security fix to prevent retrieval of the password for the SMTP, LDAP, and Proxy setup.
Version 7.4.2
ETS version 7.4.2 adds support for running multiple T-server on distinct GPUs when the host is configured with more than one GPU. This update is relevant for ETS deployments that utilizes neural language pairs.
- When a T-server is started up, it will automatically search for the first unused GPU, and if found, will mark it as reserved and proceed to use it.
- If a T-server is started up and all GPUs are in use, it will share one of the GPUs that are in use based on a round-robin algorithm.
Version 7.4.1
ETS version 7.4.1 is a minor release that resolves a previously known limitation with version 7.4.0:
- Neural MT Language Pairs now support additional mime-types besides plain-text (text/plain). Rich text format and other mime-types tags placements may not be preserved fully, but translation of these file types will not longer result in translation error.
Version 7.4.0
ETS version 7.4.0 is a significant release that introduces support for:
- Neural Machine Translation (NMT) Language Pairs
- SDL Neural MT provides a powerful framework to overcome some of the tough challenges that face Machine Translation. For example, with SDL Neural MT, we observe an impressive 30% improvement on English-German, a language pair that was historically challenging for the SMT technology. Unlike previous SMT models which are mainly limited to surface text, SDL Neural MT uses a deep learning architecture capable of learning the meaning of the text which enable the machine to perform the translation task at a semantic level leading to fluent and naturally sounding translation output. SDL Neural MT captures also both local and global dependencies and can handle long-range word reordering leading to better translation quality between languages of different word order.