Pre-Translation

Pre-translation helps speed up the translation process and ease the work of translators. It also allows you to save funds as part of translations can be automated. You can use and combine the following types of pre-translation:

• AI Pre-Translate - uses AI Models to pre-translate the content.
• Machine Translation (MT) Pre-translate – takes translations from supported machine translation engines (MTE) and applies them to the project content. You need to configure at least one of the translation engines to enable this feature.
• Translation Memory (TM) Pre-translate – takes translations from the TM vault uploaded to the system and applies them to project content.

Tip

Use the Pre-translation Accuracy report to test which pre-translation methods work the best for your content.

Crowdin AI Read more about Crowdin AI and how to use it.

Machine Translation Read more about Machine Translation and how to use it.

Translation Memory Read more about Translation Memory and how to use it.

Manual Pre-translation

To apply Pre-translation manually, follow these steps:

1. Open your project and select Dashboard on the left sidebar.
2. Click Pre-translate and select the preferred method (i.e., via Translation Memory, via Machine Translation, or AI).
3. Configure the following pre-translation parameters:
   • Target languages – select the languages to pre-translate into.
   • Files for pre-translation – select the files or folders you want to pre-translate.
   • Apply to untranslated strings only – when selected, pre-translation will only be applied to strings that don’t have any translations.
   • Allow duplicate translations – adds translations even if they duplicate existing ones. Useful when comparing AI prompts or MT engines in the Pre-translation Accuracy report, as it ensures that all translations are registered as new and included in accuracy calculations.

     Note

     Apply to untranslated strings only and Allow duplicate translations are mutually exclusive. Only one of them can be selected at a time.

   • Filter by labels / Exclude labels – include or exclude specific strings based on their assigned labels.

     Read more about Labels.

   • Minimum match ratio (Specific to pre-translation via TM) – set a similarity threshold between 40% and 100%, or select Perfect match (101%).

     Read more about TM Match Types.

   • Skip approved translations (Specific to pre-translation via TM and MT) – prevents overwriting strings that already have approved translations.
   • Approve added translations (Specific to pre-translation via TM) – allows you to auto-approve translations added through pre-translation. Options include:
     • All – approve all added translations.
     • With perfect match – approve only those with a 101% match.
     • With perfect match (approved previously) – approve perfect matches only if they were already approved before.
     • All (skip auto-substituted translations) – approve everything except translations improved by auto-substitution.
   • MT engine (Specific to pre-translation via MT) – select the machine translation engine to use for pre-translation.
   • AI prompt (Specific to pre-translation via AI) – select or create the prompt that will guide the AI model’s behavior during pre-translation.
4. Click Pre-translation.

Note

If the TM Auto-substitution feature is enabled when pre-translating via TM, it means that TM suggestions improved to 100% will be treated as regular 100% matches.

Pre-translation Queue

The Pre-translation queue allows you to trigger multiple manual pre-translations via TM, MT, or AI, each with different settings. After initiating a pre-translation, each instance enters a queue with a status of Pending.

Upon triggering a manual pre-translation, you will receive a notification indicating that the pre-translation has been added to the queue. To review all pre-translations currently in the queue, navigate to Tools > Pre-translation queue in your project. Alternatively, click Pre-translation in the project’s Dashboard and select Queue to be redirected directly to the Tools > Pre-translation queue.

Each entry in the Pre-translation queue table includes the following details:

• Created at: Displays the date and time when the pre-translation was initiated
• Type: TM, MT (MT engine name), AI (AI prompt name)
• Status: Pending, In progress, Completed, Canceled

You can cancel any pre-translation that is still in the queue if needed. Additionally, for completed pre-translations, you can view a report detailing the target languages, number of translations added and skipped, file list, and more.

Automatic Pre-translation

When working with projects with a workflow, you can add AI Pre-translation, MT Pre-translation, or TM Pre-translation workflow steps to your project workflow to automate the Pre-translation of your content.

Read more about Workflows.

Handling QA Issues During Pre-translation via AI

If the AI provider returns translations with QA issues during pre-translation, Crowdin Enterprise will collect the strings to which these faulty translations were applied and automatically reinitiate the pre-translation. The affected strings will be sent back to the AI provider along with details of the detected QA issues to ensure that the translations are corrected and meet quality standards.

Translation Memory Priority During Pre-translation via TM

When using pre-translation via TM and the Share Translation Memories option is enabled, Crowdin Enterprise applies translations from translation memories based on a specific priority order. This priority ensures that translations come from the most relevant source, optimizing translation consistency and quality.

The order of priority is as follows:

1. Default TM: Matches are first searched for in the project’s default TM.
2. Assigned TMs: If no match is found in the default TM, the system continues searching through any TMs assigned specifically to the project.
3. Shared TMs: If no match is found in the default or assigned TMs, translations are retrieved from shared TMs available to the project.

This sequence allows Crowdin Enterprise to apply translations from the most relevant TM source first, ensuring that high-quality matches are prioritized when multiple TMs are available.

Q&A

My project workflow includes a Switch Source Language step (e.g., switching from Tamil to English) followed by the TM Pre-translation step. Which source language will be used for automatic and manual pre-translation?

When a Switch Source Language step is included in the workflow, the behavior of pre-translation depends on how it’s triggered.

• Manual pre-translation will always use the initial project source language (Tamil in this case).
• Workflow-based pre-translation (e.g., an automatic TM pre-translation after the source language switch) will use the switched source language (English).

This applies to TM, MT, and AI pre-translations. To use English as the source, rely on the workflow’s automatic pre-translation step after switching the source language.