A travel company wants a voice and text booking assistant that recognizes intents and slot values, asks follow-up questions when details are missing, and calls its reservation APIs. Which pairing fits best?
Show answer and explanation
Answer: B
Amazon Lex provides automatic speech recognition and natural language understanding to identify intents and slot values. It manages multi-turn dialog, prompting for missing information. A Lambda fulfillment function can call the reservation APIs and return results to the bot. Both services scale without server management. Polly synthesizes speech from text, Connect is a contact center service, Comprehend analyzes text for entities and sentiment, Transcribe converts speech to text, and Kendra searches documents.
Why the other options are wrong
- A. Polly produces speech from text and does not understand intent. Connect is a contact center service, so this pairing cannot interpret booking requests or fulfill them.
- C. Comprehend extracts entities and sentiment from text, and Step Functions orchestrates workflows. Neither one gives a conversational bot that gathers slot values across turns.
- D. Transcribe converts speech to text and Kendra searches indexed documents. Neither provides intent recognition with slot prompting, so the bot could not drive the booking dialog.