Reports on machine translation quality
We are expanding on the recently introduced support for the new Language Weaver plugin with Machine Translation Quality Estimation (MTQE) capabilities by introducing a new report that presents an analysis of projects using such capabilities.
When you use a translation provider that offers the Quality Estimation (QE) functionality, such as Language Weaver Cloud, you can generate reports at the project level that indicate the quality of the translated segments. This new report provides valuable insights into the quality of machine translations used in your projects, enabling you to make informed decisions and optimize your translation workflows based on the MTQE data.
The reported data is divided into four categories: Good, Adequate, Poor and Unknown. The breakdown of these categories is available under the NMT scoring breakdown table within the report.
If you want the report to include locked segments, ensure that you enable the Trados Studio setting Report locked segments as a separate category. You can find instructions for configuring this setting in the Trados Studio documentation: Specifying batch settings: Analyze Files.
To access the new report, go to Projects view > Switch to Reports view and select MTQE Analysis from the Reports drop-down menu:
Known limitation
The MTQE report is not supported for the Mid-Project Update flow in GroupShare 2020 CU12.