v2.0.0 April 7, 2026
AI at Your Voice
Transform any text with a spoken instruction, ask questions mid-recording, and enjoy a completely redesigned interface — Yapper 2.0 is the biggest update yet.
AI Transform
- Select text in any app, speak a voice instruction, and get an AI rewrite — all without leaving the keyboard
- Results stream in with Markdown rendering into a glassmorphic overlay card
- Works system-wide in any application
"Hey Yapper" Voice Q&A
- Say “Hey Yapper” during any recording followed by a question to get an AI-powered answer
- Responses stream into a floating overlay right where you’re working
- 35+ wake phrase variants recognized for natural, flexible activation
7-Day Free Trial
- Fully functional free trial with no credit card required
- All features unlocked for 7 days so you can experience the full power of Yapper
- Tamper-resistant with Keychain-based storage
Microphone Selection
- Choose your preferred input device directly in Settings
- Hot-plug support with automatic fallback when a device is disconnected
- Live audio level meter for instant visual feedback
Redesigned Settings
- macOS System Settings-style NavigationSplitView with a sidebar layout
- New dedicated AI pane for provider, model, and API key management
Glassmorphism UI
- Frosted glass overlays using ultraThinMaterial throughout the app
- New design tokens system for consistent theming
- Spring animations for AI response cards
- Fully dark/light mode aware
Breaking changes
- Auto-enhancement (automatic AI post-processing) has been removed. AI is now strictly user-initiated via Transform and Q&A modes.
v1.3.0 February 21, 2026
Supercharged Transcription
A second engine, real-time streaming, and up to 190x real-time speed — Yapper has never been this fast.
Multi-Engine Transcription
- Added NVIDIA Parakeet as a second transcription engine alongside OpenAI Whisper
- Parakeet TDT v3 supports 25 EU languages at 110-190x real-time speed
- Parakeet TDT v2 available for English-only with highest accuracy
- Models are grouped by engine in the Settings picker
- Parakeet TDT v3 is now the default model for new installations
Streaming Transcription
- Text now streams in word-by-word as soon as the model begins transcribing (AI enhancement off)
- No more waiting for the full transcription to complete — results appear progressively as they’re generated
Model Storage Management
- See how much disk space your downloaded models are using directly in Settings
- Clear all downloaded models with one click to free disk space
- Cancel in-progress model downloads at any time
Expanded Language Support
- Parakeet v3 brings 25 EU languages including Bulgarian, Croatian, Czech, Danish, Estonian, Finnish, Greek, Hungarian, Latvian, Lithuanian, Maltese, Romanian, Slovak, Slovenian, and Swedish
- Language picker now filters automatically based on the selected engine’s supported languages
Bug fixes
- License activation now displays specific error details from the server instead of generic messages
v1.0.0 January 13, 2026
Initial Release
We’re excited to introduce Yapper — a native macOS dictation app that brings fast, private, on-device voice-to-text transcription to your Mac. Powered by OpenAI Whisper and optimized for Apple Silicon.
On-Device Transcription
- Voice-to-text transcription powered by OpenAI Whisper, running entirely on your Mac’s Neural Engine
- Choose from 5 Whisper models: Tiny, Base, Small, Large v3, and Large v3 Turbo (recommended)
- Automatic audio resampling to 16kHz mono for optimal transcription quality
- Custom vocabulary support for specialized terminology
Multi-Language Support
- Transcribe speech in 40+ languages including English, Spanish, French, German, Japanese, Chinese, and many more
- Easy language switching from the Settings panel
Global Hotkey Activation
- Start recording from anywhere with customizable global keyboard shortcuts (default: Cmd+Shift+Space)
- Smart detection: quick press for toggle mode, hold for push-to-talk
- Cancel anytime with Escape
- Direct text injection into any focused application via macOS Accessibility APIs
Optional AI Enhancement
- Bring your own API key to enhance transcriptions with AI-powered grammar correction, formatting, and clarity improvements
- Support for multiple providers: OpenAI, Anthropic, Google Gemini, and xAI
- Fully customizable enhancement prompts to match your writing style
- API keys stored locally on your Mac — never sent to Yapper servers
Audio File Transcription
- Transcribe existing audio files in MP3, WAV, and M4A formats
- Drag-and-drop or file picker selection
- Progress tracking with the ability to cancel mid-transcription
- One-click copy of transcription results
Transcript History
- Browse and search all past transcriptions in a dedicated history window
- Copy, delete, or bulk-manage transcription records
- Automatic cleanup of old records (configurable retention period)
- View source metadata for file transcriptions
Visual Recording Overlay
- Floating status indicator shows recording state with animated waveform visualization
- Color-coded states: recording (red), processing (blue), enhancing (purple), success (green)
- Non-intrusive design that doesn’t steal focus from your work
Privacy First
- All speech-to-text processing happens locally — your audio never leaves your device
- AI enhancement is optional and only sends transcribed text (not audio) to your chosen provider
- Transcript history stored locally with automatic cleanup
- Menu bar app that stays out of your Dock
System Requirements
- macOS 14.0 (Sonoma) or later
- Apple Silicon Mac (M1/M2/M3/M4) recommended
- 2GB free disk space for Whisper models
