Fail fast, fail cheap, making voice apps the guerrilla way
summary
Argues for rapid guerrilla voice prototyping that gets the dialogue 90% right before code and shows a working method: a one-hour Wizard of Oz test using Say Wizard — a tiny tool of 15 lines of Python that reads scripted prompts and lets a human simulate Alexa. Demonstrates a live test that records a one hour meeting and routes the file, reports adoption at BBC EDF and KLM, and stresses iterating dialogs to reveal hard problems like names and context.
AI-written summary · about
chapters (5)
related talks
32mPrototyping Voice Experiences: Design Sprints for the Assistant
19mAlexa, show me a good voice UI
50mDesigning Experiences For Voice Interactions
44mBuilding HAL: Running Ruby with Your Voice
44mDelivering On The Promise Of Voice With Alexa
44mDeveloping Skills & Actions for Voice Assistants
29mCEO of VoiceFlow on the Importance of the Voice Community
18mFacilitating Development of Pragmatic Competence through a Voice-driven Video Learning Interface