(((VERSION HISTORY)))
Release Notes
The complete version history of SonaLab for Mac, iPhone, iPad, and Apple Watch. Current version v0.8.0.
v0.8.0LatestmacOSiPhone / iPad
2026-10-12
- Register terminology overhaul: the "TA/CT Balance" chart is renamed "Fold Balance", with states unified as thick vibration / balanced / thin vibration across charts, help docs, and the glossary
- Whistle support (first batch): pitch detection limit raised to 1600 Hz; dedicated feature cards and badge for the whistle register (above 1000 Hz), plus a whistle tier on the register card
- Detection threshold calibration refresh: series of calibrations from multiple reference recordings (vibrato, breathiness, glottal closure, resonance depth), including male whistle validation
- Help docs and chart FAQs expanded: larynx position, singer formant, resonance depth, glottal closure, TA/CT meter, waveform, loudness, onset detection
v0.7.2macOS
2026-10-01
- Serial AI-chat reliability fixes: region availability gating, cold start, save notifications, settings dialogs, and window interactions
- Top-bar and window visual refinements
v0.7.1macOS
2026-09-29
- Local MCP server (tech preview): external tools such as AI coding clients can call SonaLab acoustic analysis directly — batch analysis, structured reports, and live "sing-along" monitoring; off by default, enable as needed
- AI assistant settings fix: reopening the dialog no longer overwrites custom model names with preset defaults
- Reference-recording calibration refresh: threshold corrections for vibrato, breathiness (NAQ), glottal closure (GNE), fold balance (H1-H2), and onset quality
- 15 security and privacy hardening items: injection defense, path confinement, log redaction, in-app account deletion (App Store compliance)
- Chart help fix: the "?" button no longer shows "no document" and opens real documentation
v0.7.0macOSiPhone / iPad
2026-08-17
- Cloud LLM for the AI assistant (bring your own API key): multiple provider presets and custom endpoints across OpenAI-compatible / Anthropic / Gemini protocols
- Native tool calling for the AI assistant: symptom details, targeted exercise recommendations, and a vocal-concept knowledge base; first-run setup guide added
- Full Chinese / English bilingual UI on Mac and iPhone / iPad: settings, library, student profiles, AI, charts, timeline, and playback
- Microphone calibration infrastructure with SNR gating: automatic reminders in noisy environments and visible measurement confidence
- Chart gear buttons jump straight to their settings; model picker gains catalog search
- App Store compliance preparation plus many UI detail fixes (login, sidebar, secondary navigation, warm-up / cooldown screens, etc.)
v0.6.8macOSiPhone / iPadApple Watch
2026-07-31
- iPhone / iPad edition: core charts and symptom detection, immersive-mode recording, sidebar navigation, theme unified with the Mac app
- Apple Watch edition: on-wrist real-time pitch monitoring (note name, cents deviation, frequency) and symptom summary
- DSP calibration: series of corrections to vibrato thresholds, voice-onset scoring, creak veto, nasality levels, and harmonic-noise metrics
- Performance: large drops in memory use for long-recording merge / re-analysis, spectrum GPU buffer reuse, timeline caching improvements
- AI assistant no longer stutters on long replies; re-analysis moved off the main thread so the UI never freezes
- Licenses now tolerate 7 days offline; fixed Keychain writes and sign-in edge cases; sensitive fields redacted from debug logs
v0.6.7macOS
2026-05-31
- Fixed several P0 DSP algorithm defects
- Fixed data corruption when merging recordings
v0.6.6macOS
2026-05-31
- Stability maintenance release (checkpoint before the DSP overhaul)
v0.6.5macOS
2026-05-02
- User-interface polish
- Sign-in and licensing flow improvements
- Immersive mode updates
- New AI assistant section in Settings
v0.6.4macOS
2026-04-30
- Full debugging pass
- Immersive mode improvements
- Brand-new app logo
- Fixed an exit crash
v0.6.3macOS
2026-04-30
- Routine release: feature and documentation sync
v0.6.2macOS
2026-04-22
- Local AI assistant upgrade: native llama.cpp bridge with Gemma model support
v0.6.1macOS
2026-04-11
- Architecture and experience improvements: more robust error handling, faster startup
- UI module split and refactor
v0.6.0macOS
2026-04-01
- Start of the V0.6 series
- Fixed microphone-permission click-to-trigger issue
- Enabled Hardened Runtime
v0.5.5macOS
2026-03-19
- Added sign-in alerts and subscription UI
- Unified app-wide visual theme (single SonaLab theme + VI colors)
- Systematically throttled high-frequency UI updates, eliminating performance warnings
- Fixed microphone permission issues
- Structured logging (OSLog) and Metal GPU acceleration groundwork
v0.5.4macOS
2026-02-24
- Extended license validity
- New test suites
- Direct-edition DMG build pipeline
- Unified UI theme
- Added MIDI export
- Preset management system
v0.5.3macOS
2026-02-21
- Code-quality and localization refinements
- Unified license terminology
- Large-file split optimization
v0.5.2macOS
2026-02-20
- Documentation cleanup and standardization
v0.5.1macOS
2026-02-20
- Cloud model registry and prompt sync
- Test infrastructure
- AI feature enhancements
- Device binding, account system, and pitch octave offset
v0.4.7macOS
2026-02-13
- Pitch accuracy display with smart note-name labels
- New resonance-depth trapezoid chart
- Heatmap and playback bar layout refinements
- Improved chart contrast in the light theme
v0.3macOS
2026-02-07
- Core analysis suite took shape: unified timeline recording & playback, side-profile view, L0/L1 chart system
- 15 DSP analysis modules and 8-category symptom auto-detection
- Student profile management
- Breath management training: maximum phonation time (MPT) tracking, source-filter visualization
- Vocal health system: vocal dosage tracking, cool-down recovery sessions
- Smart coach and resonance tuning indicators
- Annotation preset templates and A/B range comparison
- Page-turn follow playback; bilingual Chinese-English interface