Exam-style
A contact centre records every call. The recordings need to become text. The transcripts then need checking for personal data and scoring for sentiment. A summary of each call has to be readable by a Spanish-speaking team in Madrid. And a voice bot on the front door should handle the three most common requests, such as checking a delivery date, before anyone reaches a human. Which AWS managed AI services do those four jobs, in that order?
Reveal the answer
C. Amazon Transcribe, Amazon Comprehend, Amazon Translate, Amazon Lex
Each job has a purpose-built service. Audio into text is Amazon Transcribe. Analysing text that already exists, both the personal-data check and the sentiment score, is Amazon Comprehend. Moving text between languages is Amazon Translate. A front door built from defined intents and slots, with “when is my delivery” as an intent and the order number as a slot, is Amazon Lex. A Lex bot speaks its replies in an Amazon Polly voice, so Polly is in the finished pipeline but is not one of the four jobs. The first distractor swaps the two speech services and the two analysis services. That swap is worth rehearsing: Polly turns text into speech, Transcribe turns speech into text, and here the conversion runs from recording to transcript. Amazon Rekognition analyses images and video; it reads text inside a picture, not a transcript. The second distractor puts Amazon Quick, the enterprise AI assistant, on the front door. Quick answers open-ended questions grounded in company data, a different architecture from three defined intents wired to delivery tracking. Amazon Bedrock covers the middle three, though not with one model across all four: audio needs a speech model such as Amazon Nova Sonic, and text output varies between runs. Amazon SageMaker AI means data, labelling, training and an endpoint of your own, for four jobs that already have a managed service.
Q. Calls to text, transcripts checked for personal data and sentiment, summaries into Spanish, and a voice bot handling three common requests. Which services, in order?
A. Amazon Transcribe, Amazon Comprehend, Amazon Translate, Amazon Lex, with Amazon Polly speaking the bot’s replies.
Why? The direction of the conversion separates the pair that gets swapped most: audio to text is Transcribe, text to audio is Polly. Comprehend analyses text that already exists and returns what is in it (entities, personal data, sentiment) rather than producing new text. Lex handles a fixed set of intents and slots, a narrower job than an assistant answering open questions grounded in company data. Amazon Bedrock could cover the middle three, though the audio would need a speech model such as Amazon Nova Sonic rather than the same model prompted again, and training your own in Amazon SageMaker AI turns four API calls into four build projects.