AutoML Translation release notes

This page documents production updates to AutoML Translation. We recommend that AutoML Translation developers periodically check this list for any new announcements. Major changes will also be announced via the [google-translate-api](https://groups.google.com/forum/#!forum/google-translate-api) mailing list.

You can see the latest product updates for all of Google Cloud on the Google Cloud page, browse and filter all release notes in the Google Cloud console, or programmatically access release notes in BigQuery.

To get the latest product updates delivered to you, add the URL of this page to your feed reader, or add the feed URL directly.

October 28, 2022

The following language translation pairs have been added:

Language Pair Language Codes
Javanese -> English jv -> en
Khmer -> English km -> en
Myanmar (Burmese) -> English my -> en
Nepali -> English ne -> en
Pashto -> English ps -> en
Tagalog (Filipino) -> English tl -> en
Tamil -> English ta -> en
Telugu -> English te -> en
Uzbek -> English uz -> en
Zulu -> English zu -> en

April 03, 2020

Integration with VPC Service Controls is now in beta stage.

October 14, 2019

AutoML Translation is now generally available. Translations are now faster and quality should be improved or on par with previous versions; individual results will vary depending on your datasets. In addition, quality improvements can vary across language pairs.

Model lifetime AutoML Models run on a managed platform that is regularly upgraded. AutoML Models need to be re-trained at least every 18 months to ensure consistent performance and stability in translations over time and to get any benefits of the upgrades since original training. Models created before AutoML GA (for example, models created during service Alpha /Beta) need to be re-trained within 18 months of GA to move all assets to the newest version of the platform.

AutoML Translation has upgraded its Google NMT base models onto a new training architecture so that it now runs end-to-end on upgraded TPUs. Going forward, all training using Google NMT as the base model trains and serves on TPUs.

Customers already using AutoML might experience additional differences between v1 and previous versions of the service:

  • Data validation now provides warnings on data import to indicate certain data is not usable for training jobs.
  • Data import takes longer in order to support data validation.
  • To achieve higher quality on the new architecture, training durations are longer.

September 04, 2019

AutoML Translation has enabled Google Cloud Platform (GCP) Console-based UI for all its customers. This improves the Cloud AutoML Translation experience by adding UI elements such as controls to sort/filter datasets, which provides consistency with the rest of GCP and allows for faster release of new features.

The URL for the AutoML Translation UI has changed to https://console.cloud.google.com/translation.

September 18, 2018

AutoML Translation supports many new languages as source or target languages paired with English, bringing the total number of supported language pairs to 100. See Language Support for Custom Models for the complete list.

August 24, 2018

The URL for the AutoML Translation UI has changed to https://cloud.google.com/automl/ui/translation. The previous URL https://beta-dot-custom-vision.appspot.com/translation/ will remain active.

August 02, 2018

AutoML Translation supports Swahili (sw) as a target language from English (en) as a source language.

Language Pair Language Codes
Swahili <- English sw <- en

July 24, 2018

Models created during the EAP will continue to be valid for the Beta. You do not need to recreate/retrain models. However, the format of resource IDs (dataset IDs, model IDs, and operation IDs) is changing. If you are using hardcoded resource IDs, you need to obtain new IDs for those resources (using the get or list method) by August 31, 2018.