ElevenLabs: docs-first dialect corrections + our own v3 fixture set - #5
Merged
Conversation
… own v3 fixtures Aligning the pre-v3 dialect to the ElevenLabs docs rather than the speechmarkdown-js reference (which never had heavy testing): - break durations normalize to the documented seconds format (250ms -> 0.25s) and clamp to the 3s limit - expressive audio tags ([laugh], …) are dropped in the pre-v3 dialect: they are eleven_v3-exclusive and pre-v3 models read them aloud as literal text - new elevenlabs-v3-* corpus fixtures authored by us (the JS generator has no v3 platform) and enforced by the corpus runner via .elevenlabs-v3.ssml; submodule bumped to elevenlabs-v3-fixtures
The runner change was missing from the previous commit — as pushed, nothing read the new fixtures. Also pushed the submodule commit to feat/expand-test-fixtures (the branch .gitmodules tracks) so the gitlink stays reachable from the tracked branch.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Follow-up to #4, applying the "support v3 as fully as we can" principle and deliberately preferring ElevenLabs' documentation over speechmarkdown-js parity (the JS reference never had heavy testing and predates v3):
250ms→0.25s, still clamped to the 3s limit. The JS-generated corpus fixtures used ms; updated them (submodule bumped to theelevenlabs-v3-fixturesbranch of speechmarkdown-test-files).elevenlabs-v3-*corpus cases with.elevenlabs-v3.ssmlfixtures authored against the Rust formatter — not generator output, since the JS reference has no v3 platform. The corpus runner enforces the new extension; suite is now 217/217.Together with AACTools/rust-tts-wrapper#35 (with-timestamps degrade fallback), this closes the no-API-key-needed items from the v3 fullness review. Remaining live items need
ELEVENLABS_API_KEY.Test plan
cargo clippy --all-features --all-targetsclean