Workflow stages

By default, the projects include the task Translation. If you need to translate a document for internal use, for example, an office memo, letter or announcement, this will be enough.

For the translation of technical, legal or medical texts containing complicated terms and requiring the accurate conveyance of complicated content, you should add Editing to the workflow.

Editing is also necessary for translations of official documents, advertising materials, company websites, or any other texts intended for publication.

Ideally, you should also add Proofreading to the workflow for additional spelling and grammar checks.

To add a task to the workflow, click on it or drag it into the list.

To delete a task, hover the cursor over it and click the trash bin icon.

To delete all the tasks, press Reset.

To set up workflow quickly, you can use templates. Simply click a suitable template or drag it onto the list of tasks.

To create your own template, set a task sequence and click Save template.

 If the project includes more than one task, the documents should be processed sequentially by whoever you assigned to handle each task. This way, those parts of the text that have already been translated immediately become available to the editor, and those that have been edited - to the proofreader. Thus, the assignees do not need to wait until the previous stage is completed to begin their assigned tasks. In addition, there is no need to send the work files back and forth: they are stored on the Smartcat server and are always available to the workflow participants.

You can set up and run Pretranslation.

When you set up pretranslation, the system inserts TM matches and machine translation into the text before the work is started. This significantly speeds up the translation process, because the translator will not have to copy translation memory matches or machine translation into each segment manually.

Let's suppose you have enabled on the project both a TM that has already been used to record translations, and machine translation.

Now, on the Workflow stages page, click the Pretranslation switch.


Inserting Numbers from the Source

If your text contains a lot of numbers, click Source (Numbers Only) so that all the segments containing only numbers are copied to the translation). Decimal and thousand separators in this case will be converted to match the norms of the target language.

You can specify that the system must confirm such segments right away.

The rule for inserting numbers is automatically executed first.


Inserting Translation Memory Matches

Press Add Rule.

Click Translation Memory.

Select which TM you want to use from the ones you have enabled on the project.

In the field Insert if a match is no less than specify the minimum percentage of coincidence between a segment in the document and the translation memory at which the system must insert a translation from the TM.

You can also specify the minimum length of a segment (in words) at which the matches must be inserted using the Insert if a segment has no fewer words than field.

Finally, specify in the Confirm segments field if the system should confirm the inserted translations.

If you are sure of the quality of the translations in the selected TM and have set the minimum match percentage at 100%, you can specify that the system must confirm such matches. If you have doubts about the quality of the TM or have set the match threshold below 100%, you are better off not specifying automatic confirmation of such matches. In this case matches will be inserted, but the translator will check them, making corrections if necessary, and then manually confirming them.

If the project workflow includes several stages, you can choose at what stage segments are to be confirmed. For instance, if you know that only high-quality translations have been saved to the translation memory, you can specify that the system must confirm exact matches directly at the editing stage.

 By the same token, you can set up rules for each TM in a project (if there is more than one). The rules are executed one after the other; therefore you should set up rules whereby matches from the most reliable translation memory are inserted first.

You can set up more than one rule for each TM: for example, you can have Smartcat first insert 100% matches and confirm them, and then insert matches that are 75% and up, but without confirmation.


Inserting Machine Translation

You can set up and run pretranslation later, but keep in mind that it will only be applied to those segments that have not yet been translated.

After configuring pretranslation, press the Finish button.



Was this article helpful?