Speak To Me: Voice App Conversation Practices
summary
Covers concrete voice-app practice: use platform NLP (Alexa/Lex, Google) with service-driven architecture, handle slots and session state, and add SSML for tone. Cites 8M+ Echo installs, Google showing 70% voice input and a worrying 3% retention, and Domino’s gaining 4% market share in Q4 2017 by ordering “my regular.” Demos include a SignalR WebSocket conference search built in 48 hours, Lcars skill, smart‑home and flash‑briefing patterns, and practical auth and UX choices that reduced friction.
AI-written summary · about
chapters (12)
- 0:00Start
- 13:07Grice's Maxims
- 21:15Use Cases
- 28:26FLASH BRIEFING SKILLS
- 29:55SMART HOME SKILL
- 30:51CUSTOM SKILL
- 32:04VOICE ACTION SKILL
- 32:34INVOCATION NAME
- 33:04PORTAL CONFIGURATION
- 41:05ONE TO MANY RELATIONSHIP
- 41:48SSML Example
- 43:47ADDITIONAL REFERENCES
related talks
45mKeynote: Conversations with Machines
48mTen Things Every Voice Application Should Do
52mThe most natural UI. Building voice-enabled experiences with Amazon Alexa
10mEmbracing the Voice Revolution
19mAmazon wants to talk to you
33mStrategies for Conversational Product Experiences
52mGoing Full Duplex - Building Apps for Speech
47mBuilding Voice-First Android Apps