Examkin

Question

Difficulty: HardAWS Machine Learning Services

A media company is building an automated metadata archiving pipeline for its catalog of Spanish-language news broadcasts. The pipeline must perform the following tasks:
1. Convert the spoken Spanish audio of the broadcasts into written Spanish text.
2. Translate the generated Spanish text into English text.
3. Analyze the video frames to detect and extract on-screen news banners (text) and identify prominent public figures or brand logos.
To minimize custom machine learning model development, the company wants to use pre-trained AWS AI services. Which combination of AWS services should the company use to meet these requirements?

  1. Amazon Transcribe to convert the spoken audio to text, Amazon Translate to translate the text, and Amazon Rekognition to extract text and identify public figures from the video framesAnswer
  2. B
    Amazon Polly to convert the spoken audio to text, Amazon Comprehend to translate the text, and Amazon Rekognition to extract text and identify public figures from the video frames
  3. C
    Amazon Transcribe to convert the spoken audio to text, Amazon Translate to translate the text, and Amazon SageMaker to build custom models to extract text and identify public figures from the video frames
  4. D
    Amazon Lex to convert the spoken audio to text, Amazon Translate to translate the text, and Amazon Rekognition to extract text and identify public figures from the video frames

Answer

The combination of Amazon Transcribe to convert the spoken audio to text, Amazon Translate to translate the text, and Amazon Rekognition to extract text and identify public figures from the video frames.
The correct option correctly maps each task to the corresponding fully managed, pre-trained AWS AI service: Amazon Transcribe converts the spoken Spanish audio to text, Amazon Translate translates the Spanish text into English, and Amazon Rekognition extracts on-screen text and identifies public figures from the video frames.

Step-by-Step Solution

1
Identify the service required to convert the pre-recorded spoken Spanish audio into written text.
Amazon Transcribe is selected because it is the AWS service designed for speech-to-text conversion.
Converting spoken audio to written text requires a transcription service, not a conversational chatbot (Lex) or a text-to-speech converter (Polly).
2
Identify the service required to translate the transcribed Spanish text into English text.
Amazon Translate is selected because it is designed for fluent text translation between languages.
Translation is the primary function of Amazon Translate, whereas Amazon Comprehend is designed for NLP analysis such as sentiment and key phrase detection, not language translation.
3
Identify the service required to analyze the video frames to detect text and recognize public figures/logos.
Amazon Rekognition is selected because it is a computer vision service with pre-trained capabilities for text-in-image extraction (OCR), object detection, and celebrity recognition.
Using Amazon Rekognition avoids custom model development, unlike Amazon SageMaker which would require building and training models from scratch.

Key Concept

AWS Machine Learning Services
Estimated Time:2m 0s
Rate this question

Topics

AWS Machine Learning Services

More questions from this topic