Smartcat has various tools like translation memory, machine translation, and glossaries that help you improve the quality of translations. So lets quickly overview each of them.
Inserting translations from the CAT panel
The CAT panel displays all the linguistic assets applied to the project for a segment selected: matches from the translation memory, machine translation, and glossary terms.
For each entry, the CAT panel shows the source text, its translation, and the asset type: a translation memory match (TM), machine translation (MT), or a glossary term (TB).
Additional information about a CAT panel entry is displayed in the CAT info panel below.
To use a particular entry, double-click on it in the CAT panel, and it will be inserted into the Target field of the current segment.
To insert the translation (by default, the first entry is selected in the CAT panel), you can also press Ctrl+Space. It's easy to switch between entries on the CAT panel, use Ctrl+Up or Ctrl+Down then press Ctrl+Space to insert the entry.
The other option is to press Ctrl plus the number of the translation or an entry in the CAT panel (Ctrl+1, Ctrl+2, etc.).
Translation memory matches
When you are working on a segment, the system looks through the translation memories enabled in the project. If the system has found a match, the translation will be shown within the CAT panel.
By the way, the percentage threshold from which entries of the translation memory will appear in the CAT panel can be set in the Linguistic Assets tab of the project.
Translations from the TM are sorted in the descending order of the match percentage.
The system analyses the similarity of TM entries with a segment and then displays the match percentage. If a TM match is 100%, this means that the entry and the segment are the full-text matches. If a full-text match from the translation memory differs from the segment by only one numeral, the match is still considered a 100% match while the system replaces the numeral from the match to the one in the source segment.
The system checks not only the text within a segment but also the context—that is, the previous and next segments. If one of the adjacent segments of the segmented selected matches the TM entry context, it is considered a 101% match. If both adjacent segments match the context of the TM entry, it is considered a 102% match.
The more you work in the system, and the more often similar or identical sentences appear in documents you translate, the more time you save by inserting previously translated entries accumulated in the TMs.
TM consistency control
Translation memories allow not only saving time and efforts while translating similar texts, but they also ensure that consistency is met.
If a full-text or context match is found, but your translation differs from that match, a warning of a possible mistake will appear in the QA Check tab.
A warning also appears if the segment has several exact or context matches. In this case, you can compare the available translations and verify that you have selected the best option.
Another check prevents accidental confirmation of a fuzzy match. If you insert a match that is less than 100% and confirm the translation without changing anything in it, the QA Check tab will display a warning. It's up to you to make a change to the segment or ignore the warning.
For more info about error messages see Quality Assurance.
Whenever you translate contracts, technical descriptions, quotations, or other standard documents, machine translation can save you huge amounts of time. In most cases, machine translation will require editing. However, by using it as a draft, you will complete your task much faster, especially if you set up pretranslation for the project.
Please, keep in mind that machine translation is provided by third-party suppliers - that's why it is a paid service except for the Yandex engine, which collects impersonal feedback to improve its quality. You can manage additional services on the Productivity Services menu.
Whether a machine translation engine could be used for a language pair depend on the supported languages of such an engine.
The CAT panel displays the terms available for the selected segment from the glossaries enabled to the project. Thus, you don't have to manually search for the terms in glossaries.
The terms are also highlighted in yellow in the source text. If you select a term in the CAT panel, it is highlighted in orange.
To insert a term from the CAT pane to the target field, simply press Ctrl plus the number of the term in the CAT pane (Ctrl+1, Ctrl+2, etc.).
You can add terms into glossaries applied to the project either on the glossary page or directly in the Editor.
If you are working on a project that a client has created in the system, you might not have the right to work with glossaries, but you can still suggest a term. In this case, the client will see the suggestion then add it to the glossary, edit, or reject it.
To add or suggest a term, press or Ctrl+E. Then fill in the required fields and press Add.
The term will be immediately added to the glossary and will then be available in the CAT panel.
As opposed to translation memories where only one TM is selected for writing, it is possible to add (or suggest) terms to any of the glossaries associated with the project.
To edit a term, select it in the CAT pane, and in the CAT info pane click Edit.
Make changes you want and press Save.
You can also work with terms on the glossary page. Click Go to term in the CAT info panel. The glossary will open in the new browser tab.
To learn more about glossaries, see Creating a Glossary.
Glossaries not only help linguists employ the appropriate terminology while translating, but they also ensure the consistency of the terminology in your texts. The system checks the translation of each segment, and if it finds that the translation of a term does not match the glossary, a warning will appear in the QA Check tab.
If you use a term marked as not recommended, the system will warn you of an error.
Forbidden terms are marked red in the CAT panel.