Microsoft Azure AI Fundamentals (AI-901)

When deploying a generative AI model as a service, applications interact with it via a defined API endpoint. This endpoint is a URL that specifies where the model service can be accessed. To secure access and ensure only authorized applications can call the model, authentication credentials (such as an API key, access token, or managed identity) are required. Therefore, an application must include both the model's endpoint URL and the necessary authentication credentials to successfully send requests and receive responses from the deployed model.

Named Entity Recognition (NER) is the appropriate text analysis technique for identifying mentions of specific entities like people, organizations, and locations within text. It classifies these entities into predefined categories, directly addressing the requirement to automatically identify such mentions in customer support emails. Sentiment analysis determines emotional tone, summarization condenses text, and key phrase extraction identifies main topics, none of which directly fulfill the task of classifying named entities.

To align with the Microsoft Responsible AI principle of inclusiveness, AI applications should be designed to empower everyone, regardless of their background or abilities. Designing an interface to support multiple languages and screen readers directly addresses this by ensuring the application is accessible and usable by a wider, more diverse audience, including individuals with visual impairments or different linguistic preferences. This promotes equitable access and engagement.

Option A pertains to transparency, as it explains how predictions are generated. Option C relates to fairness, aiming to mitigate bias across demographic groups. Option D is an example of privacy and security, protecting customer data and restricting access.

✅ Option A (Correct)

Reasoning: The scenario requires building an AI solution that "produces new product images based on written descriptions provided by users." This directly describes the functionality of image generation (also known as text-to-image generation or generative AI for images). This AI workload takes textual input (descriptions) and synthesizes corresponding visual content (new product images).

❌ Why the other choices are incorrect:

• Option B is incorrect: Image analysis involves processing and understanding existing images to extract information (e.g., classification, content moderation, feature extraction). It does not create new images.
• Option C is incorrect: Object detection is a specific form of image analysis that identifies and locates instances of objects within an image. It does not generate images from text.
• Option D is incorrect: Optical character recognition (OCR) is used to extract text from images or scanned documents. It has no role in generating new visual content.

The scenario requires converting written customer notifications into natural-sounding spoken audio for a phone system. This process is known as Text-to-Speech (TTS).

Speech synthesis, also known as Text-to-Speech, is the Azure AI Speech capability specifically designed to transform text input into human-like synthetic speech. It allows customization of voice, language, and speaking style, making it ideal for applications like automated phone systems, voice assistants, and accessibility features.

Speaker recognition identifies who is speaking. Speech recognition (Speech-to-Text) converts spoken audio into text. Speech translation translates speech into another language, often involving both speech-to-text and text-to-speech or direct speech-to-speech, but its core function is translation, not just synthesizing text to speech.