XTM Cloud 12.5

Date: October 11, 2020

General

Linguists or users with a limited PM role can now be set up automatically via SSO.

XTM now supports a selection of African languages: Bemba, Changana, Dholuo, and Kasena; languages spoken in northeastern India: Manipuri and Nagamese; and indigenous languages of Americas: Kaingang and Nheengatu.
It is now possible to bulk import users from an Excel file with the information about the workflow steps assigned to them.
XTM rate cards now support four decimal points in price entries.
Contractors can now decide whether LSPs can see the name of the project manager assigned to the project.
The rate card settings related to minimum charges have been improved. It is now possible to define a minimum charge at project, language and task level.
Managing the workflow for hourly-based projects is now easier. It is now possible to define the estimated time spent on a task and propagate it across all files at language level.

The HTML parser has been enhanced to process inline tags in XTM more accurately.
XTM can now accurately process .po files that handle plural forms in different languages.

A new QA check is now available in XTM Workbench. It generates a warning if the translation for a fuzzy match has been inserted into a target segment without being edited.
A new option has been added to XTM Workbench: auto-propagation can now apply to filtered repeated segments only.
Linguists working in XTM Workbench can now easily copy segment IDs with just one click. It is also possible to hide segment IDs while still benefiting from the ID-based search and filtering options.
The XTM Workbench docked panel has been redesigned for an optimal user experience. The tab can be now opened in a new browser window.
XTM Workbench users can now choose where they wish to perform the Concordance search: in approved translation memory entries only or in all of the translation memories available.
Full metrics can now be viewed directly from within XTM Workbench.

With the use of Inter-Language Vector Space, a proprietary technology which maps relationships between words across languages, the XTM linguistic AI features such as aligner, bilingual terminology extraction, and automatic placement of inline tags have been improved to support 10 additional Indian languages: Assam, Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Nepali, Panjabi, and Sindhi.
The AI-driven bilingual term extraction feature has been further enhanced. The extraction process can now determine and eliminate generic words from the generated dictionary based on the criteria of minimum frequency and Term Frequency – Inverse Document Frequency (TF-IDF) weighting.

A new project workflow step has been provided to downgrade XLIFF:doc statuses “Proofed” and “Validated” to “Translated” for ICE and Leveraged matches offered in Translation Memory.

XTM Cloud now provides the option to enable and configure the connection to a custom MT engine.

A new REST API method allows users to send Purchase Orders to linguists and the PO creator by email.
Project tags can now be defined per source file when creating projects and updating source files using REST API.
The REST API has been improved to allow creating and updating projects with locale-specific source files.
Using a new REST API method, you can now reanalyze projects when uploading new files as well as at any time in the workflow process.
The TAUS DQF integration now offers an additional ‘metadata only’ API type that allows end-users to work with DQF analytics without sending the translation content to TAUS. The type of data to be sent to TAUS can be selected at system, customer, project template and project level.
There is a new XTM REST API method to activate or archive a project.