Building Conversational Experiences with Actions on Google
summary
Overview of Actions on Google with concrete building blocks: the Assistant does speech-to-text and uses Dialogflow converts text to intents and extracts entities and contexts to drive flow; required slots become prompts, with a three‑try disambiguation policy. Backends connect via a webhook that exchanges JSON — webhook receives requests and returns conversation responses (use the JavaScript client or any JSON parser). Supports SSML and pre-recorded audio, suggestion chips and cards for phones, account linking, push updates, and four voices per language. Transactions roll out in Germany next year.
AI-written summary · about
chapters (20)
- 0:00Introduction and overview
- 1:03Why Google Assistant matters
- 5:59Actions on Google explained
- 12:01Live demo of a conversational app
- 14:37How the Google Assistant works
- 17:11Dialogflow for natural language
- 19:02Designing intents and entities
- 25:57Connecting to backend services
- 28:44Creating a great app experience
- 30:30Designing conversational personas
- 33:00Handling user input and errors
- 34:09Advanced speech markup and audio
- 34:51User persistence and personalization
- 35:54Designing for multiple surfaces
- 38:05User updates and notifications
- 39:29Account linking and transactions
- 41:55Discovering your app
- 43:23Getting started with Actions Console
- 45:10Workshop and resources
- 46:16Q&A session
related talks
40mOk Google, we need to talk
15mSession 5: Fireside chat with the Google Assistant DevRel team
34mOk Google! We need to talk
45mExtending the Google Assistant with (Actions on Google and) API.ai
34mBuilding Conversational Apps with Firebase and Google Actions
52mGoogle Assistant VS Alexa: Battle of the Bots
46mChatbots, shall we shift into second gear?
51mMake Your Gadget Talk: Google Assistant