Section 23: Machine Learning Flashcards
(23 cards)
What can Amazon Rekognition detect in images and videos?
Objects, people, text, and scenes using machine learning.
What facial features can Amazon Rekognition analyze?
Gender, age range, and emotions.
What are some use cases of Amazon Rekognition?
Labeling, content moderation, text detection, face detection and analysis, face search and verification, celebrity recognition, and pathing.
How does Amazon Rekognition verify users?
Through facial analysis and facial search.
Can Amazon Rekognition detect celebrities?
Yes, it includes celebrity recognition.
What is the purpose of the ‘familiar faces’ database in Amazon Rekognition?
To compare detected faces against known individuals or celebrities.
What is ‘pathing’ in Amazon Rekognition used for?
Sports game analysis.
What does Amazon Rekognition’s content moderation feature detect?
Inappropriate, unwanted, or offensive content in images and videos.
In what industries is Amazon Rekognition content moderation commonly used?
Social media, broadcast media, advertising, and e-commerce.
What user experience goal does Amazon Rekognition content moderation help achieve?
It helps create a safer user experience.
What can you set in Amazon Rekognition to determine which content gets flagged?
A Minimum Confidence Threshold.
What tool can be used for manual review of flagged content in Amazon Rekognition?
Amazon Augmented AI (A2I).
How does Amazon Rekognition content moderation assist with legal compliance?
It helps comply with regulations.
What is amazon transcribe
Converts speech to text
This converts text to speech
Polly
In Polly what to use to customize the pronunciation of words
Lexicons
In Polly what can you use to
* emphasizing specific words or phrases
* using phonetic pronunciation
* including breathing sounds, whispering
* using the Newscaster speaking style
Speech Synthesis Markup Language (SSML)
What is amazon connect
Virtual contact center
This service is for Natural Language Processing(NLP)
Comprehend
This service is used for people to build ML models
SageMaker
Fully managed document search service powered by Machine Learning
Amazon Kendra
Fully managed ML-service to build apps with real-time personalized recommendations
Amazon Personalize
Automatically extracts text, handwriting, and data from any scanned
documents using AI and ML
Amazon Textract
