You are building an agent that needs to create a summary of a 10-minute audio recording. Which workflow is most efficient?
check_circle
Correct AnswerB
Transcribing audio to text first is the standard pattern for enabling an LLM to reason over and summarize spoken content.
Question 32
Which Azure AI Search component is responsible for extracting text from a collection of Word and PDF files stored in Azure Blob Storage?
check_circle
Correct AnswerD
The Indexer is the crawler that connects to the datasource, opens the files, and extracts content for the index.
Question 33
You want to use a "Small Language Model" (SLM) for a simple classification task on a local device. Which model is the best fit?
check_circle
Correct AnswerA
Phi-3 is a Microsoft-developed SLM designed for high efficiency and a small compute footprint.
Question 34
Which metric in AI Foundry Evaluation measures the "Similarity" between two vectors to determine how well the search result matches the query?
check_circle
Correct AnswerA
Cosine Similarity is the standard metric for measuring the orientation (semantic similarity) between two high-dimensional embeddings.
Question 35
Which Azure service is used to host a "Custom Skill" for an Azure AI Search enrichment pipeline that uses a specific proprietary library?
check_circle
Correct AnswerB
Azure Functions provides the serverless execution environment needed to run custom code as part of the search indexing process.
Question 36
You need to ensure that your AI-103 application logs every prompt and completion for auditing purposes. Which Azure service should you use to store and query these logs?
check_circle
Correct AnswerB
Log Analytics is the primary workspace for collecting, indexing, and querying diagnostic data from Azure AI services.
Question 37
You need to monitor the "Token Consumption" of your Azure OpenAI deployment to stay within budget. Which Azure service provides the metrics and dashboards for this?
check_circle
Correct AnswerA
Azure Monitor tracks diagnostic metrics including ProcessedTokens and GeneratedTokens for OpenAI resources.
Question 38
You are evaluating an agent's response for "Relevance". What specifically are you comparing?
check_circle
Correct AnswerC
Relevance measures how well the model's response addresses the user's intent as expressed in the initial prompt.
Question 39
You want to implement "Rate Limiting" to ensure that no single internal department exceeds their allocated AI budget. Where should this be implemented?
check_circle
Correct AnswerD
API Management provides the tools to set throttling, quotas, and rate limits on a per-key or per-group basis.
Question 40
Which Azure AI service provides the ability to translate text from one language to another while preserving the "Tone" and "Context"?
check_circle
Correct AnswerB
Azure AI Translator is the dedicated service for high-quality text-to-text translation across multiple languages.
