docs: correctness fixes + em-dash style sweep (atoms + waves) - #359
Draft
abhishekmishragithub wants to merge 3 commits into
Draft
docs: correctness fixes + em-dash style sweep (atoms + waves)#359abhishekmishragithub wants to merge 3 commits into
abhishekmishragithub wants to merge 3 commits into
Conversation
…uplicated page - speech-to-text/pre-recorded/audio-formats: "No limit on file size" was wrong; the pre-recorded STT API caps requests at 250 MB (larger returns HTTP 413), per the STT OpenAPI spec. - voice-cloning/how-to-vc-api: cloned voices run on lightning_v3.1 only; cloning is not available on lightning_v3.1_pro (per the TTS spec). Removed the Pro option from the compatibility table. - llm (overview, quickstart, supported-parameters): "Migrate from OpenAI" card/ link pointed at migrate-from-open-ai, which does not exist; the page slug is migrate-from-openai. Fixed 3 references. - deep-dive/realtime-events/events: the entire page body (plus a stray inline copy of the frontmatter) was duplicated. Removed the duplicate, keeping one copy of each section and the trailing callout.
Mechanical sweep: replaced the em-dash character with a hyphen across all MDX prose in both product doc trees (house style: no em-dashes). Code fences are left untouched, and en-dash numeric ranges (e.g. 5–15 seconds) are preserved. 1702 replacements across 288 files. Substitution is mechanical; a few sites may read better re-punctuated (comma/colon/parentheses) on a later editorial pass.
|
🌿 Preview your docs: https://smallest-ai-preview-chore-docs-style-correctness.docs.buildwithfern.com |
Earlier this file was edited to claim cloning is unavailable on lightning_v3.1_pro (matching the TTS/cloning specs). The platform actually supports Pro cloning, so the spec is stale, not the doc. Restored the row to list both lightning_v3.1 and lightning_v3.1_pro. Spec enum + SDK regen for the exact create-clone Pro model token are a separate follow-up (need the runtime token confirmed against the API).
Collaborator
Author
|
Correction: the earlier "cloned voices are lightning_v3.1-only" change was reverted — the platform supports cloning on |
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.
Two commits, reviewable separately.
1. Correctness fixes (spec-verified)
audio-formats.mdxsaid "No limit on file size"; the API caps requests at 250 MB (larger → HTTP 413), per the STT OpenAPI spec.how-to-vc-api.mdximplied clones run onlightning_v3.1_pro; the TTS spec says cloning islightning_v3.1only. Removed the Pro option.migrate-from-open-ai(nonexistent). Correct slug ismigrate-from-openai. Fixed 3 refs.deep-dive/realtime-events/events.mdxhad its entire body (and an inline copy of the frontmatter) pasted twice. Removed the duplicate, kept one copy + the trailing callout.2. Em-dash style sweep (mechanical)
5–15 seconds) preserved.Coverage audit result (no code change)
Ran a parallel SDK-vs-spec-vs-docs coverage audit alongside this. SDK feature coverage is complete for both the current release and 6.0 (analytics, billing, call-actions, concurrency, agent-versioning branch model, webhooks incl.
update, TTS/STT/voice-cloning/electron/streaming all present). An initial "missing endpoints" scare traced to auditing a stale local working tree, not a real gap.Genuinely open items (not in this PR):
speech_started/speech_ended) aren't modeled as typed receive messages instt-live-ws.yaml, so the SDK can't surface them typed. Spec under-modeling.docs.jsonorphan in the atoms tree (repo ships on Fern).pulse.mdxstreaming-vs-pre-recorded language counts need a careful spec reconcile (left untouched — couldn't verify cleanly).Draft — for review, not auto-merge.