Hybrid glottal source - Finally fixed the "breathy" voice quality at 16000 Hz and above! The synth now picks the right waveform per sample rate: smooth cosine at low rates (no aliasing), LF-inspired buzz at 16kHz+ (that Eloquence-like crispness).
VoicingTone v2 - 3 new DSP params:
• Noise glottal modulation (Klatt-style pulsing)
• Pitch-sync F1/B1 modulation (vocal tract coupling)
All with backward-compatible version detection so old drivers keep working.
Phoneme Editor - Full voicing params UI! 10 sliders, per-voice storage, auto-loads when you switch voices. Plus 3-tier DLL fallback for older builds.
NVDA Addon - New voicing sliders in synth settings ring. Tweak brightness, tilt, crispness per-voice.
Linux builds - Fresh x86_64 .so files included.
This was a marathon session - glottal waveforms, struct versioning, crash fixes, UI work... mass refresh across the whole stack.
github.com/tgeczy/NVSpeechPlay…
github.com/tgeczy/NVSpeechPlay…
github.com/tgeczy/NVSpeechPlay…
Joshua
in reply to Bri🥰 • • •Bri🥰
in reply to Joshua • • •Joshua
in reply to Bri🥰 • • •Kyle Smith
in reply to Bri🥰 • • •Bri🥰
in reply to Kyle Smith • • •Kyle Smith
in reply to Bri🥰 • • •