Project scope
v0.1 goals
GlyphFlow has one browser-native job: speak English text while filling each spoken word continuously from left to right inside its letters.
| Area | v0.1 scope |
|---|---|
| Speech | Browser Web Speech API (speechSynthesis) |
| Caption | A continuous CSS-gradient wipe inside each word, guided by word boundaries |
| Integration | Framework-independent TypeScript DOM API; safe to import during SSR |
| Initial target | Desktop Chrome, English text, voices that emit word boundaries |
Deliberate boundaries
| Area | Outside this library's scope |
|---|---|
| Audio pipeline | Audio files, MP3 generation, and forced alignment |
| External services | Servers, network requests, and external AI models |
| Data & analytics | Does not save data or collect analytics |
| Compatibility | A guarantee for every browser, OS, voice, or language |
| Voice quality | Improving naturalness, voice availability, or emitted boundary behavior |
Voice quality and boundary behavior come from the browser and operating system, not GlyphFlow.
Development commands
npm install
npm run lint
npm run format:check
npm test
npm run build
npm run demo
npm run docs:build