Conversational UIs: Talking to Siri, Alexa, and Your Web Browser
summary
Demos show concrete claims and working code: browser text‑to‑speech needs two lines and uses Web Speech API (speechSynthesis), richer pronunciation uses SSML, and recognition can use JSpeech Grammar Format with confidence scores. Gartner’s 30% voice interaction prediction appears as motivation. Examples include a Roku voice search that actually finds Game of Thrones, an Alexa “space geek” skill wired to Lambda returning SSML, and an 11‑year‑old typing to Siri so it speaks her school report — recognition still relies on cloud, noise and accents remain challenges.
AI-written summary · about
chapters (19)
- 0:00Conversational UIs intro
- 1:23Why voice is essential
- 3:30Voice as extra interaction
- 7:06Conversational vs traditional UIs
- 8:14History of conversational UIs
- 10:13Modern text-to-speech tech
- 12:31Accessibility in voice UIs
- 13:35Real-world impact on accessibility
- 16:46Speech synthesis in browser
- 19:14Advanced SSML synthesis
- 21:42Speech recognition challenges
- 23:40Context in voice interactions
- 26:27Cloud speech limits
- 27:53Browser speech demos
- 28:57Voice command grammars
- 31:06Testing Alexa skills
- 31:48Building Alexa skill
- 34:19Future of conversational UIs
- 37:56Q&A
related talks
48mConversational ULS: Talking to Siri, Alexa, and Your Web Browser
58mAdding Speech to your Web App_UIArchitecture
49mDesigning voice interfaces
19mAlexa, show me a good voice UI
16mWeak signals of a vocal revolution in UIs
44mDeveloping Skills & Actions for Voice Assistants
21mBuilding a Voice-Enabled AI Assistant With Javascript
20mVoice games – An overview of an up-and-coming category