Developing Skills & Actions for Voice Assistants
summary
Concrete how-to: builds a tiny “Namesayer” conversational app, deploys it as a Namesayer demo to Google Assistant via Dialogflow + Actions console and as an Alexa custom skill backed by a Lambda. Shows that most conversation design requires only intent/parameter definitions and randomised responses, while backend work (smart‑thermostat, device control) lives in cloud functions and must meet strict latency limits. Google integration sometimes misparses live input, Alexa’s SSML made responses feel more natural, and Google’s prebuilt types are being deprecated.
AI-written summary · about
chapters (9)
- 0:00Introduction and setup
- 2:32What are voice assistants
- 8:16Designing for voice interactions
- 23:48Technical implementation overview
- 26:59Demo introduction
- 27:21Google Assistant demo walkthrough
- 31:54Alexa demo walkthrough
- 36:11Wrap-up and future outlook
- 39:02Q&A session
related talks
20mBuilding Voice Apps for the Google Assistant
44mDelivering On The Promise Of Voice With Alexa
56mGoogle vs Alexa: Battle of the Bots
48mConversation Actions for Google Assistant
52mGoogle Assistant VS Alexa: Battle of the Bots
17mOK Google: Build Interactive Apps with Google Assistant
41mVoice and Visual: How to Build Actions for Smart Displays and More
33mBuilding Conversational Experiences with Actions on Google