[MLS-C01] [IMPL and OPs] Recommend and implement

Recommend and implement the appropriate machine learning services and features for a given problem

Posted by Oscaner on August 7, 2022

Appropriate services

Select from several AWS Services to implement the appropriate Machine Learning solution for your business problem

Lex

  • Build conversational interfaces into your applications using voice and text
  • ML using deep learning algorithms of automatic speech recognition (ASR) for converting speech to text, and natural language understanding (NLU) to recognize the intent of the text
  • Use the same deep learning algorithms used in Alexa to build natural language conversational chatbots
  • Fully managed service that scales automatically

Rekognition

  • Add video and image analysis to your applications
  • ML using deep learning algorithms to identify objects, people, text, scenes, activities, and inappropriate content
  • Facial recognition and facial search for verification, people counting, celebrity identification, public safety
  • Rekognition Custom Labels feature automatically identifies and labels in your video or images

Transcribe

  • Add speech-to-text to your applications
  • ML using deep learning algorithms such as automatic speech recognition (ASR) to convert speech to text
  • Can process speech in batch or near real time in a streaming application
  • Can automatically redact content such as personally identifiable information (PII)

Translate

  • Add language translation to your applications
  • Neural machine translation service for language translation that uses deep learning models to produce accurate, natural sounding translation
  • Localize content
  • On-demand translation
  • Continually learning from expanding datasets

Polly

  • Add text-to-speech (TTS) to your applications
  • Uses deep learning algorithms to synthesize human speech that sounds natural
  • Has many voices/languages from which to choose
  • Also has Neural Text-to-Speech (NTTS) voices which support newscaster and conversational styles
  • Can create a brand voice for your organization, a customized NTTS voice

Comprehend

  • Natural language processing (NLP) service that finds insights and relationships in text
  • Identifies language, extracts key phrases, places, people, events
  • Understands positive or negative sentiment of the text
  • Tokenizes text and parts of speech and organizes text files by topic
  • Can extend to recognize your own organization’s vocabulary

本文由 Oscaner 创作, 采用 知识共享署名4.0 国际许可协议进行许可
本站文章除注明转载/出处外, 均为本站原创或翻译, 转载前请务必署名