Machine Translation

MTU Consumption and Weights

Content is machine translated from English by Phrase Language AI.

Monitor MT usage in the MT usage tab of Phrase Language AI.

MTU Calculation

Phrase calculates MTU usage based on each character sent to a machine translation service using the following calculation:

A * (B + C) = Machine Translation Units

Where:

  • A is the total number of characters that the customer sends for machine translation.

  • B is the weight assigned by Phrase to the applicable machine translation service.

  • C is the weight assigned by Phrase to reflect the additional features made available in Phrase Language AI (e.g. autoselect, QPS, MT glossaries, etc.).

    This weight is always applied to any machine translation performed via Phrase Language AI, even if the additional features are not used.

Features that consume MTUs are:

  • Pre-translation

  • Default analysis

    MTUs are consumed when:

    • The option Include machine translation matches (QPS) is enabled.

    • There are no 100/101% TM or non-translatables (NT) matches.

      There is a prioritization mechanism in case of TM, NT and MT matches with the same score. TM matches have the top priority, while MT matches come last.

      If the Default analysis finds a 100% or 101% TM match, MT or NT matches are not requested. Similarly, if there are no 100% TM matches and the analysis finds a 100% NT match, MT is not requested.

    • Files are uploaded to a Submitter Portal that is set up to use MT.

  • CAT editor

  • Phrase Portal

Examples

Manually managed engine with autoselect and QPS:

  • 10 characters [A] * (0 [B] + 0.5 [C]) = 5 MTUs

Manually managed engine with autoselect and without QPS:

  • 10 characters [A] * (0 [B] + 0.5 [C]) = 5 MTUs

Fully managed engine (e.g. Phrase NextMT) with autoselect and QPS:

  • 10 characters [A] * (0.5 [B] + 0.5 [C]) = 10 MTUs

Fully managed engine (e.g. Phrase NextMT) with autoselect and without QPS:

  • 10 characters [A] * (0.5 [B] + 0.5 [C]) = 10 MTUs

Machine Translation Service Weights

Type

Weight

Amazon Translate

0.5

Custom NextMT

0.5

DeepL

0.5

Google Translate

0.5

Microsoft Translator

0.5

Phrase Next GenMT

1.0

Phrase NextMT

0.5

Rozetta Translate

2.0

Tencent

0.5

Phrase Language AI Service Weights

Type

Weight

With or without additional features

0.5

Optimizing MTU usage

  • Use the Apply MT only when there's no TM match at or above the threshold in pre-translation to ensure that MT is only used when there is no TM match at or above the set threshold.

  • When translating a job in the editor, use the Suggest MT only for segments with a TM match below {set value}.

    This ensures:

    • MT suggestions will only be offered if there is no TM match at or above the set threshold.

    • Autocomplete will not use MT units to make suggestions.

    • MT is not suggested in following workflow steps for segments confirmed in the previous step, as there will be an adequate TM match available.

  • When creating a default analysis, disable the Include machine translation matches (QPS) option.

Was this article helpful?

Sorry about that! In what way was it not helpful?

The article didn’t address my problem.
I couldn’t understand the article.
The feature doesn’t do what I need.
Other reason.

Note that feedback is provided anonymously so we aren't able to reply to questions.
If you'd like to ask a question, submit a request to our Support team.
Thank you for your feedback.