docs: document remote dictation worker engine

[omnigent-ci]

Documentation update for omnigent-ai/omnigent#3025 — feat(dictation): remote worker engine for offloading speech-to-text

Changes documented

• app/docs/deploy/dictation-worker/page.mdx — new page documenting the remote dictation engine and standalone dictation worker: engine selection (OMNIGENT_DICTATION_ENGINE), running the worker (python -m omnigent.server.dictation_worker), the unauthenticated-worker security warning, pointing the main server at it via OMNIGENT_DICTATION_REMOTE_URL, a config-var reference table, and per-take fallback behavior.
• components/DocsSidebarFull.js — added a Dictation Worker entry to the Collaboration section, next to Cloud Sandbox Host.

Manual review needed

• app/docs/deploy/dictation-worker placement — dictation had no prior docs on the site, so I created a fresh Deploy page. A reviewer may prefer folding a shorter version into app/docs/interact/web-ui (the mic button lives there) or app/docs/deploy/overview; the worker is an operator/scaling concern, so Deploy seemed the natural home.
• Base dictation feature (mic button, sherpa engine, omnigent[dictation] extra, scripts/fetch-dictation-models.sh) is still undocumented site-wide — out of scope for this PR (it predates it), but worth a follow-up so this new page isn't the only dictation reference.

---

Source PR: omnigent-ai/omnigent#3025 · merged by @kerryspchang
_{Drafted automatically by the doc-sync workflow. Review for accuracy before merging.}

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
omnigent-site	Ready	Preview, Comment	Jul 22, 2026 12:59am