v0.2.2
This release improves Apple SpeechAnalyzer transcription reliability for macOS 26+ on Apple Silicon.
- Apple Speech is limited to English, Chinese, Japanese, and Korean, including both Simplified and Traditional Chinese
- Stopping capture now finalizes the latest volatile SpeechAnalyzer result instead of discarding the final words
- SpeechAnalyzer input preserves captured-audio timestamps through format conversion
- Settings activation now retries until the window is visible and active
- Added automated checks for supported languages and audio timestamps, plus manual stop-finalization QA
This asset is ad-hoc signed because no Developer ID certificate is available in the release environment. See the README installation note if macOS quarantine blocks first launch.