Administration

Phrase AI Features Overview

Phrase includes several AI-powered capabilities across the platform, spanning translation, content adaptation, and quality evaluation.

Note

Availability varies by feature and subscription plan. For each feature's availability, detailed configuration, and usage instructions, see the dedicated articles.

AI Translation Agent

A GenAI-powered agent that produces context-aware translations with built-in review and refinement. It performs context-aware translation, enforces terminology, applies translation memory-grounded consistency checks (RAG), and corrects fluency and tag or formatting issues.

Phrase Language AI (Machine Translation)

Phrase's machine translation (MT) and agentic AI translation hub. It connects Phrase capabilities to MT engines (native, ecosystem, and bring-your-own) and manages how translation requests are configured, routed, and evaluated through MT profiles and MT Glossary support.

Custom AI

An environment for building custom machine translation engine models from translation memory data, covering dataset creation, training, and deployment. This allows organizations to build models tailored to their own content and terminology.

MT Optimize (Orchestrator)

A generative AI post-editing step in Phrase Orchestrator. It improves existing machine translation output by refining fluency, grammar, terminology, and formality.

Auto Adapt (Orchestrator)

A generative AI content transformation feature in Phrase Orchestrator. It adapts content for a target audience or tone, generates variants, and adjusts reading level.

AI Checks

Automated checks that run during quality evaluation. Each AI check comes from a rule and returns a pass or fail verdict for the content it checks. Checks are grouped automatically by content group and language into a quality profile.

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.