AWS Certified Solutions Architect – Associate (SAA-C03) — Question 490
A company wants to use artificial intelligence (AI) to determine the quality of its customer service calls. The company currently manages calls in four different languages, including English. The company will offer new languages in the future. The company does not have the resources to regularly maintain machine learning (ML) models.
The company needs to create written sentiment analysis reports from the customer service call recordings. The customer service call recording text must be translated into English.
Which combination of steps will meet these requirements? (Choose three.)
Answer options
- A. Use Amazon Comprehend to translate the audio recordings into English.
- B. Use Amazon Lex to create the written sentiment analysis reports.
- C. Use Amazon Polly to convert the audio recordings into text.
- D. Use Amazon Transcribe to convert the audio recordings in any language into text.
- E. Use Amazon Translate to translate text in any language to English.
- F. Use Amazon Comprehend to create the sentiment analysis reports.
Correct answer: D, E, F
Explanation
To meet the requirements without managing custom ML models, the company should use Amazon Transcribe to convert the audio recordings into text, Amazon Translate to translate the resulting text into English, and Amazon Comprehend to perform sentiment analysis on the translated text. Amazon Polly is incorrect because it converts text to speech, not speech to text. Amazon Lex is used for conversational chatbots rather than sentiment report generation, and Amazon Comprehend does not perform language translation.