AWS On Air ft. Automated chatbot designer in Amazon Lex
summary
Amazon Lex adds an automated chatbot designer that ingests conversation transcripts, runs ML to extract intents, slots and sample utterances, and produces a first bot design in a few hours instead of the many weeks a manual process takes. It recommends roughly 10,000 lines of transcripts or representative samples, commonly surfaces 20–30 intents for a domain, outputs associated transcripts for each intent, and lets teams import and iterate on the generated bot; preview is free.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 1:37Chatbot Designer
- 2:52Chatbots
- 4:21Automated chatbot designer
- 7:42Design is a iterative process
- 8:50Use cases
- 9:37How it works
- 10:16Lets see it in action
- 10:38Creating the bot
- 12:10Reviewing the bot
- 13:36How much data do you need
- 15:10Adding intents
- 15:35Adding context
- 17:10Modify design
- 17:30Data
- 18:32Preview
- 19:22Outro
related talks
47mBuilding sophisticated conversational AI experiences
02h 35mConversational Bots : Amazon Lex, IBM Watson, Microsoft Bot and Google Assistant
47mBuild and Deploy an Amazon Lex Chatbot and Convert it to an Alexa Skill
54melevating customer experience w/ conversational AI
27mWhat If We Built an Intelligent Voice Chatbot for Amazon Alexa in 25 Minutes?
35mDesigning Experiences for Chatbots
20mBuilding Smarter Cloud Support Systems with AWS Lex
42mConversational AI in Amazon Alexa