Phrase Language AI is Phrase's machine translation (MT) and agentic AI translation hub with additional features. It connects Phrase products to MT engines and manages how translation requests are configured, routed, and evaluated.
Where Phrase Language AI Is Used
Phrase Language AI is available across the following Phrase products:
-
Provides MT for enterprise localization workflows, including machine translation post-editing.
-
Provides MT for software and UI localization workflows, with support for MT glossaries and integration into the Strings editor.
-
Enables file-based and text translation through a secure, customizable interface using configured MT profiles and engines.
-
Provides MT for media localization workflows, including audio transcription, dubbing, and video subtitle localization.
-
Allows the deployment of the Phrase Language AI outside the Phrase Platform for high-volume MT at scale. It supports:
Configure Phrase Language AI
Important
MT profiles and engine configuration are managed in Phrase TMS by TMS Administrators and Project Managers with required permissions, even when Phrase Language AI is used in other Phrase products.
If MT profiles are enabled, engine management moves to the page, and AI Translation Agent becomes available as an MT provider in projects, templates, and Phrase Portal.
-
Team, Professional, Business, Enterprise and Software UI/UX plans
Get in touch with Sales for licensing questions.
Important
DeepL will no longer be available as a native engine from Jun 30, 2026. Enable Phrase Next GenMT or continue using DeepL via Phrase Ecosystem. If no action is taken, Phrase Next GenMT will be enabled automatically. Learn more.
An MT profile is a configuration within Phrase Language AI that contains one or more MT engines. Profiles can include any combination of native, ecosystem, and bring-your-own engines. Once an engine is integrated into Phrase Language AI, it can be enabled inside an MT profile.
An MT profile is assigned to a project in TMS, Strings, Portal, or Studio. Users can assign different MT profiles per language pair within a project.
If multiple MT engines are enabled in a profile, MT Autoselect picks the most suitable one for each job. If only one specific engine is required, either disable all other engines in the profile, or create a dedicated MT profile with only that engine and select that profile for machine translation across Phrase products.
Glossaries can be associated with MT engines that support them. When attached to an MT profile, the engine applies the glossary's terminology during translation.
Create an MT Profile
Customers on Team, Professional, Business, Enterprise, and Software UI/UX plans can create multiple MT profiles, each with a different combination of engines and settings. This allows different projects, teams, or language pairs to use different MT configurations for their projects.
To create a machine translation profile, follow these steps:
-
From the tab on the Phrase Language AI page, click Enable multiple profiles.
The table is presented.
-
Click New profile.
The window opens.
-
Provide a name for the profile and click Create.
The configuration page for the profile is presented.
-
Configure profile as required and click Phrase Language AI at the top of the page.
The new profile is presented in the tab.
Edit and Delete an MT Profile
Profiles can be edited or deleted from the menu to the right of the profile name.
Other MT Profile Actions
Profiles can be also marked as default or duplicated from the menu to the right of the profile name.
If using Phrase Language AI via API, the MT profile ID is required for translation requests and can be found in the URL of the specific MT profile page.
Example URL: https://cloud.memsource.com/tms/language-ai/profile/123abc456xyz
123abc456xyz is the MT profile ID.
MT profiles support three types of MT engines.
Native Engine
A native engine is a machine translation engine available out of the box in Phrase Language AI. It requires no contract, API key, or external setup. A single toggle activates the engine for translation.
-
Amazon Translate, also supported by Phrase Language AI via API.
-
DeepL and DeepL Next Gen, also supported by Phrase Language AI via API.
Not supported in Phrase Portal.
Important
DeepL will no longer be available as a native engine from Jun 30, 2026. Enable Phrase Next GenMT or continue using DeepL via Phrase Ecosystem. If no action is taken, Phrase Next GenMT will be enabled automatically. Learn more.
-
Google Translate, also supported by Phrase Language AI via API.
-
Microsoft Translator / Microsoft Translator Hub, also supported by Phrase Language AI via API.
-
Phrase NextMT, also supported by Phrase Language AI via API.
-
Phrase Next GenMT, also supported by Phrase Language AI via API.
-
Rozetta T-4OO Realtime, also supported by Phrase Language AI via API.
-
Tencent TranSmart, also supported by Phrase Language AI via API.
-
Widn.AI, also supported by Phrase Language AI via API.
Configure Native Engine
To enable a native engine for pre-translation:
If only one engine is enabled in the MT profile, that engine is used for pre-translation. If multiple engines are enabled, MT Autoselect selects the best engine for each job automatically.
Ecosystem Engine
An ecosystem engine is a machine translation engine configured within Phrase Language AI. It requires a separate contract with the engine provider, arranged outside of Phrase. Once the contract is in place, enter the provider's API key into Phrase Language AI to establish the connection. The engine is then available to toggle on and use for translation.
-
Alexa Translations A.I. (formerly Yappn)
-
Custom NextMT
-
DeepL
-
Google AutoML, also supported by Phrase Language AI via API.
Note
When adding Google AutoML to Phrase Language AI, Bucket name is required information due to support and potential usage of MT glossaries.
-
Human Science
-
Hunyuan
-
Lengoo HALOS
-
Microsoft Custom Translator (in Microsoft Azure, set location to "global"), also supported by Phrase Language AI via API.
-
ModernMT
Note
A ModernMT license that allows usage within CAT tools is required. To retrieve information about the license, use the
/users/meendpoint. -
NpatMT
-
Omniscien MT V6
-
Systran PNMT
-
Ubiqus NMT
-
Yandex
Configure Ecosystem Engine
To connect an ecosystem engine using an API key:
-
From the Phrase Language AI MT profile page, click Connect MT engines.
The page is displayed.
-
Select an engine.
The page for the selected engine opens.
-
Provide the API Key.
-
If required, provide any additional information for the engine.
-
Click Save.
The engine is now available for activation.
Bring Your Own Engine
Bring Your Own (BYO) Engine is a framework for connecting an MT engine that is not available as a native or ecosystem engine. Phrase provides a schema for building an adapter that connects the external engine to Phrase Language AI.
Refer to Bring Your Own Engine for full configuration steps.
Note
Formality level configuration is available for engines that support it, currently Phrase Next GenMT and DeepL.
MT Autoselect is an AI-powered feature that selects the optimal engine for each translation job, based on the job's content domain and language pair. It only considers engines that are toggled on in the MT profile.
MT Autoselect supports domain-based recommendation for all documents in these source languages:
-
English
-
Czech
-
German
-
Spanish
-
French
-
Italian
-
Japanese
-
Korean
-
Dutch
-
Russian
-
Swedish
-
Chinese Simplified
All other language combinations receive MT Autoselect recommendations but do not take text domain into account. Engine selection for these pairs is based on language pair performance only.
The type of content translated with machine translation can significantly affect the quality of the output. An engine that performs well for product descriptions might struggle with a medical report. The content type variable is known as a domain.
Categorizing documents into domains is normally a process that requires human input, creating complications when it comes to scaling translation and allowing for basic human error.
An AI model detects domains automatically, without manual labeling. The model identifies 11 distinct content types based on characteristic keywords:
|
Domain |
Keywords |
|---|---|
|
Medical |
‘study’, ‘patients’, ‘patient’, ‘treatment’, ‘dose’, ‘mg’, ‘clinical’ |
|
Travel and Hospitality |
‘km’, ‘hotel’, ‘guests’, ‘room’, ‘accommodation’ |
|
Business and Education |
‘team’, ‘business’, ‘work’, ‘school’, ‘students’ |
|
Legal and Finance |
‘agreement’, ‘company’, ‘contract’, ‘services’, ‘financial’ |
|
Software User Documentation |
‘click’, ‘select’, ‘data’, ‘text’, ‘view’, ‘file’ |
|
Consumer Electronics |
‘power’, ‘battery’, ‘switch’, ‘sensor’, ‘usb’ |
|
User Support |
‘please’, ‘email’, ‘account’, ‘domain’, ‘contact’ |
|
Cloud Services |
‘network’, ‘server’, ‘database’, ‘sql’, ‘data’ |
|
Industrial |
‘mm’, ‘pressure’, ‘valve’, ‘machine’, ‘oil’ |
|
Software Development |
‘value’, ‘class’, ‘type’, ‘element’, ‘string’ |
|
Entertainment |
‘game’, ‘like’, ‘get’, ‘love’, ‘play’, ‘go’ |
Quality Performance Score (QPS) is an AI quality estimation system that predicts the score a translation would receive if reviewed by a human using the MQM (Multidimensional Quality Metrics) framework. It takes the source text and translated target text and returns a score from 0 to 100.
QPS is enabled by default for all engines, with no separate setup required.