Screen Reader Dictionary Builder

Build and export pronunciation dictionaries for NVDA, JAWS, and Apple VoiceOver — one workspace per course (CHEM 113, CHEM 116, etc.).

Live app (after Pages is enabled): https://jordan77-lang.github.io/screenreader/ → dictionary-builder.html

One-time: turn on GitHub Pages

1. Open github.com/jordan77-lang/screenreader/settings/pages
2. Under Build and deployment, set Source to GitHub Actions
3. Go to Actions → Deploy to GitHub Pages → Re-run all jobs

Alternatively, set Source to Deploy from a branch, branch main, folder / (root) — no Actions required.

Quick start

1. Open the Dictionary Builder.
2. Choose a class tab (or add a new class).
3. Import — upload CSV/TSV/XLSX, or Load Sample From Workspace.
4. Edit terms (search, scroll panel / full list / collapsed view).
5. Export — Apple CSV, JAWS TSV, NVDA .dic, or NVDA add-on (requires NVDA 2026.1+ for add-on speech dictionaries).

Optional: shared editing (Supabase)

See BACKEND_SETUP.md and supabase/README.md.

Do not commit Supabase anon keys; paste them in the browser only.

Docs

• DICTIONARY_WORKFLOW.md — NVDA add-on export and testing
• BACKEND_SETUP.md — Supabase collaboration

Source files

Course templates and NVDA regex live under Screen reader/ (e.g. DSL_chemistry_Apple_VoiceOver_pronunciation_source_cH_gmol_no_mol.csv).

Local preview

py -m http.server 8080

Then open http://localhost:8080/dictionary-builder.html.