docs: document local TTS on macOS, Linux, and Windows#110230
docs: document local TTS on macOS, Linux, and Windows#110230ivan-digital wants to merge 6 commits into
Conversation
|
Codex review: needs maintainer review before merge. Reviewed July 18, 2026, 8:46 AM ET / 12:46 UTC. Summary PR surface: Docs +158. Total +158 across 2 files. Reproducibility: not applicable. This PR addresses missing documentation for already available provider paths, and the linked report supplies a direct guide-content verification path. Review metrics: none identified. Merge readiness Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch. Rank-up moves:
Next step before merge
Security Review detailsBest possible solution: Merge the focused guide after normal maintainer confirmation that the cited third-party installation commands and release prerequisites remain accurate at landing time. Do we have a high-confidence way to reproduce the issue? Not applicable: this PR addresses missing documentation for already available provider paths, and the linked report supplies a direct guide-content verification path. Is this the best way to solve the issue? Yes. Documenting the existing OpenAI-compatible and local CLI provider surfaces is the narrowest maintainable solution because it avoids adding a parallel runtime integration. AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning high; reviewed against af2662f25a85. Label changesLabel justifications:
Evidence reviewedPR surface: Docs +158. Total +158 across 2 files. View PR surface stats
What I checked:
Likely related people:
What the crustacean ranks mean
Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics. How this review workflow works
Review history (5 earlier review cycles)
|
|
@clawsweeper re-review |
|
🦞🧹 I asked ClawSweeper to review this item again. Re-review progress:
|
|
Refreshed the branch against current main and updated the Windows recipe to use the published speech-core v0.0.11 package and its packaged Kokoro executable. The complete documentation gate passes on the refreshed branch: formatting and markdown lint for 721 files, MDX validation for 737 files, 6,076 internal links with zero broken links, glossary validation, and a current generated docs map. @clawsweeper re-review |
|
🦞🧹 I asked ClawSweeper to review this item again. Re-review progress:
|
Related: #110229
Release prerequisite: The macOS HTTP tab now explicitly requires the first Speech Swift release containing soniqo/speech-swift#415. It states that v0.0.22 has the server binary but not the OpenAI speech route, and directs current users to the released CLI path or a source build.
What Problem This Solves
OpenClaw users can already connect TTS to a custom OpenAI-compatible endpoint or a local executable, but the guide does not show a complete private, offline setup for macOS, Linux, or Windows. Users currently have to discover the correct provider, executable arguments, model setup, and WAV handling independently.
Why This Change Was Made
This documents the existing provider surfaces instead of adding another runtime integration. It adds copyable Speech Swift HTTP and CLI configurations for macOS, a speech-core Kokoro CLI configuration for Linux, and a speech-core Supertonic LiteRT configuration for Windows. It also documents the existing OpenAI responseFormat override required by WAV-only local endpoints.
User Impact
Users can configure fully local TTS on all three desktop platforms from one guide. The setup keeps audio inference on the host and preserves current OpenClaw provider behavior; no application code or dependencies change.
Evidence