Introduction

Following are the ML services provided by AWS:

• Amazon Recognition - find objects/people-text in images & videos
• Amazon Transcribe - speech to text
• Amazon Polly - Text to speech
• Amazon Translate - Translation from on langauge to another
• Amazon Lex - Chatbot by Auto Speech Recognition
• Amazon Connect - Contact Center Automation using ML
• AWS Comprihend - NLP & finding inferences from data
• Amazon SageMaker - Building ML Models
• Amazon Forecast - Predicting data
• Amazon Kendra - Document Search Service
• Amazon Personalize - Personalized Recommendation for each user!!
• Amazon Textract - Extracting text,handwriting,data from document using AI/ML

Amazon Recognition

• Create a DB of familiar faces or compare against celebrities
• Usecases

    - Labeling
    - Content Moderation
    - Text Detection
    - Face Detection & Analysis
    - Face Search
    - Celebrity Recognition
    - Pathing

Amazon Transcribe

• Automatically converts speech to text
• Uses a deep learning process called Automatic Speech Recognition (ASR).
• Automatically removes personally identifyable information (PII) using redaction.
• Supports multilanguage Identification for multilingual audio.
• Usecase:
  • Transcribe Customer service calls.
  • Automate close captioning & sustainablity.
  • Generate metadata for media assets to create fully searchable archives.

Amazon Polly

• Turn text to lifelike speech using deep learning.
• Allow you to create application that talks.

Amazon Translate

• Natural & Accurate Language Translation
• It alows you to localize context: such as website application for international users & to easily translate large volume of lexts effectively.

Amazon Lex

• Lex is an automatic Speech Recognition (ASR) to convert speech to text.
• Natural Language Understanding to recognize the intent of text callers.
• Help to build a chatbot.

Amazon Connect

• Recieve Calls, Create contact flows, Cloud-based contact center.
• Can integrate with other CRMs or AWS.
• No Upfront Payment, 80% cheaper than traditional contact-center-solutions.

Amazon Comprihend

• Used for NLP
• Fully Managed / Serverless Service
• Use ML to find insight & relationship in text.

- Language of the text
- Extract key phases, places, people, brands, or events
- understandable how positive/negative the text is.
- Analyzes texts using tokenization & parts of speech
- Automatically organises a collection of text files by topic.

Simple Usecases

• Analyze customer interactions (email,etc) to find what leads to a positive/negative experience.
• Create & Group articles by topics that comprehend will uncover.

Amazon SageMaker

• Fully Managed services for developers/data structure to build ML models.
• Typically difficult to do all the process in one place + provision server.
• ML Process: Predicting your exam scores.
• Historical data:

    # of years of experience in IT
    # of years of experience of AWS
    Time Spent on the course.

Amazon Forecast

• Fully Managed service that uses ML to deliver highly accurate forecast.
• Example: predict the future sales of a raincoat.
• 50% more accurate than looking at data itself.
• Reduce forcasting time from months to hours.
• Usecases:

    - Product Demand Planning
    - Financial Planning
    - Resource Planning

Amazon Kendra

Fully managed document search service powered by ML. Used to extract answers from within a document(text, pdf, HTML, PowerPoint, MS Word, FAQs, etc).

Amazon Personalize

• Fully Managed ML-service to build apps with realtime personalized recommendation.
• Example: Personalize product recommendation/ re-ranking, customized direct marketing.
• Another E.G: User brought gardening tools, provide recommendation on the next one to buy.
• Some technology used by amazon.com
• Integrate into existing websites, application, sms, email, marketing system.
• Implement in days, & not months (you dont need to build, train and deploy ML Solutions)

Amazon Textract

• Automatically extract texts, handwriting & data from any scanned document usign AI & ML.
• Extract data from forms, tables, etc.
• Read & Process types of document.
• Financial Service (Invoices) & Healthcare (Reports)