docs(install): per-platform install pages — Windows / macOS / Linux (Cycle 16)#18
Merged
Conversation
…inux (Cycle 16) Fill the three install stubs at their canonical paths (the docs landing's platform picker links them). Zero-assumptions, hand-holdy walkthrough from nothing → a talking character in the browser, in two clearly-separated tiers, identical structure across all three pages: - Tier 1 (easy on-ramp, browser voice): install Docker (Desktop on Win/Mac, Engine + Compose plugin on Linux — official docs linked), get the project (clone or Download ZIP), open a terminal, `docker compose up mash` → http://localhost:8090 → upload a .acs, Speak. - Tier 2 (optional authentic TruVoice): plain-language why-files, the THREE user-supplied files into services/voice-server/vendor/ (spchapi.exe, tv_enua.exe, sdk/include/speech.h), then ONE `docker compose up` — matches the merged Cycle 15 one-command build (Docker is the only host tool; first build slower then cached; first synth of a new phrase ~few seconds, repeats instant; novel first utterance may clip slightly). Per-platform specifics verified (not guessed): Win Docker Desktop/WSL2 + PowerShell; Mac Docker Desktop + Apple-chip vs Intel + Terminal via Spotlight; Linux Docker Engine + Compose plugin (`docker compose`, not the hyphen) + run-without-sudo (docker group) with sudo fallback. All Docker links are current docs.docker.com pages. IP rule holds: NO download links to the proprietary binaries/speech.h — sourcing defers to docs/legal-and-assets.md only. Screenshots/GIFs signposted as coming (Cycle 17), not faked. Additive accuracy (same paths, no churn): docs/README.md picker line + root README drop the "(coming soon)" on the now-live install links. Docs only — no code. CI green; link audit clean (144 relative links resolve); reviewed. Cycle doc: docs/cycles/cycle-16-install-pages.md. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Cycle 16 — per-platform install pages
Fills the three install stubs the docs landing's platform picker links — at their canonical paths (
docs/install/{windows,mac,linux}.md, no renames). Each is a zero-assumptions, hand-held walk from nothing → a talking character in the browser, in two clearly-separated tiers with an identical structure across all three pages.Spec:
docs/cycles/cycle-16-install-pages.mdTier 1 — See it run (easy on-ramp, browser voice)
Install Docker → get the project (clone or Download ZIP) → open a terminal →
docker compose up mash→ http://localhost:8090 → upload a.acs, type, Speak. Clearly labeled: uses the browser's built-in voice — the easy on-ramp, not a lesser path.Tier 2 — The authentic TruVoice voice (optional)
Plain-language why, then the three user-supplied files into
services/voice-server/vendor/(spchapi.exe,tv_enua.exe,sdk/include/speech.h), then onedocker compose up→ authentic Genie. Matches the just-merged Cycle 15 one-command build: Docker is the only host tool (no Node/pnpm); first build slower then cached; first synth of a new phrase ~few seconds, repeats instant; a novel first utterance may clip slightly (honest, minor).Verified, not guessed
docker-compose.yml+services/voice-server/Dockerfileon this branch:up mashvsup, 8090/8080, the 3 vendor files incl. thesdk/include/speech.hsubpath, no host build step.docker compose, not the hyphen) + run-without-sudo (docker group) withsudofallback. All Docker links are currentdocs.docker.compages (confirmed via search).speech.h— sourcing defers only todocs/legal-and-assets.md. (Grep-verified: the only external links are docs.docker.com + localhost.)Verification
typecheck/test/lint/format; 246 tests; Markdown prettier-ignored)..mdlinks across 68 files resolve — zero broken.code-reviewer: no blockers; re-verified commands/ports/paths against the actual compose+Dockerfile, the 3-file Tier 2 flow, per-platform correctness, link depth + page consistency, nostalgia-deletes-clean voice, and no proprietary links. Two cosmetic nits applied (removed a font-dependent Apple-logo glyph; marked the still-stubcharacters.mdreferences "coming soon").Additive accuracy (same paths, no churn): the landing's platform-picker line + dropping "(coming soon)" on the root README's now-live install links.
Out of scope (signposted): developer page (next cycle), screenshots/GIFs (Cycle 17), the
voice/*deep pages.Does not merge.
🤖 Generated with Claude Code