News & Updates
What's New
Changelog, roadmap, and development updates for TypeSay.
Latest
April 2026
TypeSay v0.1.0 — Initial Release
The first public release of TypeSay. Everything you need to start dictating text privately on your desktop.
- ✓ Push-to-talk speech-to-text with Whisper AI
- ✓ Four Whisper models: Tiny, Base, Small, Medium
- ✓ System-wide global hotkey with customization
- ✓ Floating visualizer orb (Orb and Pill modes)
- ✓ Smart text insertion (type simulation + clipboard)
- ✓ GPU acceleration: CUDA, Metal, Vulkan
- ✓ 18+ languages with auto-detection
- ✓ Comprehensive settings panel (8 tabs)
- ✓ Welcome onboarding and About screen
Roadmap
What's coming next
TypeSay is actively developed. Here is what we are working on for upcoming releases.
Custom vocabulary and corrections
Train TypeSay to recognize specialized terms, names, and jargon specific to your work.
Voice commands and formatting
Say "new paragraph" or "capitalize that" to control formatting with your voice.
Whisper Large model support
Support for the Whisper Large model tier for maximum transcription accuracy on capable hardware.
Audio history and session logs
Optionally save transcription history locally for review, all stored on your machine.
Stay Connected
Follow development on GitHub.
Star the repo to get notified about new releases and features.
Star on GitHub