Microsoft Azure AI Engineer Associate AI-102 Practice Question

Each Question Answering project in Azure AI Language is bound to a single locale that determines the language used during training and querying. To serve multiple languages, you must create one project for each language and deploy each project to its own Azure resource or deployment. Your application then routes the user's utterance to the endpoint that matches the detected language, which avoids on-the-fly translation and therefore minimizes latency. Adding translation middleware or storing multilingual pairs inside one project still leaves the project monolingual and would require runtime translation, increasing latency. A single project with metadata filters is unsupported for different locales.