Skip to content

docs: link Orca stable SenseVoice integration - #355

Open
LauraGPT wants to merge 1 commit into
mainfrom
codex/orca-stable-release-20260922
Open

LauraGPT wants to merge 1 commit into
mainfrom
codex/orca-stable-release-20260922

Conversation

@LauraGPT

Copy link
Copy Markdown
Member

Summary

  • Update the English, Chinese, and Japanese Orca entries from the old prerelease-only notice to the verified v1.4.206 stable release.
  • Clarify that the SenseVoice model must be downloaded before local offline recognition, and that this integration uses non-streaming inference.
  • Update the existing README contract test to check the release link and these boundaries in each Orca entry, and reject the obsolete version notices.

User impact

Users can find a stable desktop release containing the previously merged SenseVoice integration instead of being directed to an old prerelease. No claim that v1.4.206 is the first stable release containing the integration.

Model, API, and runtime impact

  • No model behavior changes.
  • Documentation and its existing regression test only; no dependency, runtime, license, or workflow changes.

Validation

  • git diff --check
  • python -m pytest -q tests/test_whats_new.py tests/test_canonical_qwenaudio_links.py tests/test_github_templates.py tests/test_funasr_requirement.py: 11 passed.
  • python3 -m unittest tests.test_container_contract tests.test_device_env: 14 passed.
  • Checked the release metadata: published 2026-09-20, neither draft nor prerelease, with Linux, macOS, and Windows installers.
  • Compared the original integration merge with the stable tag: stable is 4,077 commits ahead, zero behind, and the merge base is the integration merge.
  • Inspected tagged model catalog and STT worker.
  • Downloaded the official orca-ide_1.4.206_amd64.deb without installing or executing it. Its SHA256 matches the release asset digest: 66e6ea0ca2aaeb706297d27bf7d6ecb8c352737954e4d527c6a66e0b1518b956.
  • Inspected the shipped app.asar: package version 1.4.206, pinned SenseVoice INT8 model-download metadata in the compiled main bundle, and the active offline SenseVoice recognizer branch in the compiled STT worker.

Notes for reviewers

The previous documentation test failed because it explicitly required the outdated prerelease URL; it passes with the synchronized contract update. The diff was self-reviewed. This is release/package-content verification, not a desktop launch, microphone, macOS/Windows execution, or new ASR accuracy/performance test. Full model and Docker build tests were not rerun locally for this documentation-only change; the repository's normal PR workflow remains enabled. No reported issue is closed by this PR.

Signed-off-by: LauraGPT <18321252+LauraGPT@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant