Ask AI

Amazon Translate with Translation Memory (beta)

Learn about the Amazon Translate with TM connector

πŸ’‘

This feature is currently in beta and may change or be discontinued as our development and product plans evolve. Availability is not guaranteed. Please contact your Account Manager or Customer Success Manager for more information on testing with this connector.

Supported languages

Amazon Translate supports some 75 languages.

Β 

Its translation memory feature, known as "parallel data," allows you to customise translations by providing your own examples of source and target language pairs. This customisation can be applied to any of the supported languages.

Enable Amazon Translate with Translation Memory

  1. Navigate to the Connectors Page
      • Click on the top-right dropdown menu in CaptionHub and select Connectors.
  1. Enable the Amazon Translate with TM Connector
      • Go to the MT Engines tab.
      • Click Edit on the Amazon Translate with Translation Memory connector.
      • Toggle the connector to enable it.
  1. Adjust optional settings (profanity, brevity, and formality)
  1. Manage Translation Memory (TM)
      • Toggle on Use imported TM.
      • Click Manage TM and follow these steps:
          1. Create a new TM by entering a name and uploading a file (CSV, TSV, or TMX format).
          1. Save the TM, then select it from the dropdown list to apply it.

Highlights

With this TM option, CaptionHub sends data to Amazon Translate using an alternative method compared to the β€˜standard’ Amazon Translate Machine Translation (MT). By incorporating the Translation Memory (TM) as parallel data, you should see improved translation results, though processing times will be longer.

For this reason, we recommend submitting translations in smaller batches.

Since the data is stored as Parallel Data, it remains within your CaptionHub instance and is not shared externally.

Did this answer your question?
😞
😐
🀩