Finding Your Voice: Building Screenless Interfaces with Node.js
summary
A practical account of building a screenless UI: a Node.js API that powers an Amazon Alexa demo, with shared backends that cover both Google Assistant and Alexa so about 60% of code is reusable, packaged via gulp and deployed to AWS Lambda (serverless is optional). SSML solves pronunciation; audio playback hits platform gaps (Google adds native audio player only Feb 2018). Surprises: no user IP/location available, QA requires manual end‑to‑end device testing, and the work is more integration and platform tradeoffs than fresh ML.
AI-written summary · about
chapters (12)
- 0:00Intro
- 1:16Why voice UI development?
- 5:29A brief timeline of voice assistants
- 6:51A natural evolution
- 8:57Conclusions
- 11:25The request/response flow
- 14:07Examples from Alexa SDK
- 15:13SSML: A common language
- 17:52Two "skills", one codebase
- 18:54Generic Response Model
- 19:37Challenges
- 23:25Open source opportunities
related talks
29mFinding Your Voice: Building Screenless Interfaces with Node.js
30mMove over GUI, let's build a VUI with JavaScript!
29mBuild and scale multiple Voice application by using TypeScript
48mBuilding Serverless AI powered Applications on AWS
01h 21mHow to Build Cross Platform Voice Apps with the Jovo Framework
57mBuilding for Alexa with Web API
55mBuilding for Alexa with Web API
21mNode & Musical Performance