Skip to content

v1.24.3

Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 13 Jun 14:46
· 1 commit to main since this release
Immutable release. Only release title and notes can be modified.
v1.24.3
1d876d7
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Bug-fix release: macOS AVSpeech voices now honor --rate, and auto-detected Chinese routes to the voice the engine actually ships.

Bug Fixes

  • --rate now works for macos-* AVSpeech voices (#548) — the speaking-rate multiplier (0.5–2.0) was silently discarded for system voices. It is now forwarded to the AVSpeech sidecar and mapped onto AVSpeechUtterance.rate, so kesha say --voice macos-… --rate 1.5 actually changes the pace.
  • Auto-detected Chinese routes to zh-zm_050 (#547) — auto-routing previously pointed detected zh at a voice the engine does not ship, failing with E_VOICE_UNKNOWN. Explicit --voice zh-zm_050 was unaffected.

Docs / Internal

  • OpenSpec baseline spec corpus (#544) — 11 capability specs capturing current behavior, with a glossary and spec-writing rules. No shipped-code change.

Platform support

Unchanged: CoreML (macOS 14+, Apple Silicon) / ONNX (macOS, Linux, Windows). Engine protocol v3.

Shipped PRs

  • #548 — fix(tts): honor --rate for macos-* AVSpeech voices
  • #547 — fix(tts): route zh auto-detection to zh-zm_050
  • #544 — docs(openspec): baseline specs for all existing features

Upgrade

bun add -g @drakulavich/kesha-voice-kit@1.24.3 && kesha install
Assets 23
Loading