In Memsource Cloud, it is possible to calculate translation memory, machine translation and non-translatables (NT) net rates. A project manager can set up one or more net rate schemes by going to Setup - Net Rate Schemes. The net rates work independently of any financial information. The number of units for translation (pages/words/characters) simply gets decreased based on the selected net rate scheme.
Translation Memory Net Rates
Translation memory net rate schemes are well established in the translation industry. These are the main categories:
- Repetition (a segment which gets repeated in a document)
- In-context match (a segment, which matches a segment already stored in translation memory, including its context = its preceding and following segment or context key)
- 100% match (a segment, which matches a segment already stored in translation memory)
- Fuzzy match (a segment, which partly matches a segment already stored in translation memory; fuzzy matches can range from 99% down to zero; however, fuzzy matches below a 75% match rate are often referred to as "new words"; i.e. they need to be translated from scratch)
Non-translatables Net Rates
- 100% match ( a segment which Memsource has identified as content that is very unlikely to require translation, but may need to be checked by a translator)
- 99% match (a segment which Memsource has identified as content that probably does not require translation, but should be checked by a translator)
- Fuzzy match - For the default analysis, NTs would fall only into one of the categories mentioned above. If these are post-edited, they can fall into the fuzzy ranges in terms of Post-editing analysis.
Working with Net Rate Schemes
When a document for translation is imported into a project, it can be analyzed (by carrying out a Default analysis) against the project's translation memories and to identify non-translatables. After opening the analysis, a net rate scheme can be selected and the pages/words/characters will get recalculated. The results will be displayed in the line Net Rate.
Typically, the Default analysis is generated before the translation.
Machine Translation Net Rates
Memsource makes it possible to measure machine translation quality by comparing machine translation and its final post-edited version at a segment level. The match score is counted in just the same way as for translation memory matches. To analyze the post-edited translation against the initial machine translation, launch the Post-editing analysis via the Analyze button under Project.
After opening a post-editing analysis, a net rate scheme can be selected and the pages/words/characters will get recalculated. The results will get displayed in the line Net Rate.
The post-editing analysis gets typically generated after the translation.