HyperYap is a local voice-to-text application that bundles speech recognition, a terminal emulator, and hotkey automation into a single zero-config installer for Windows. No cloud, no internet required for transcription, no data collection. Install once, use everywhere.
- What You Get
- Install
- Default Hotkeys
- Requirements
- How It Works
- Configuration
- Build from Source
- Contributing
- Attribution
- License
HyperYap bundles three tools into a single grab-and-go package:
| Component | What It Does | |
|---|---|---|
| HyperYap voice engine |  |
Local speech-to-text powered by NVIDIA Parakeet TDT 0.6B v3 |
| BenjaminTerm |  |
Hacker-styled WezTerm terminal with smart clipboard, 86 dark themes, and borderless mode |
| Hotkey daemon |  |
Mouse side buttons to F13 (record), CapsLock to F13, Mouse Forward to Enter. Runs as its own tray icon process. |
| Smart paste | Ctrl+V in terminals auto-saves clipboard images as PNGs and pastes the file path | |
| Auto-boot | Everything starts on login. No setup after reboot. | |
| Preset configs | Toggle-to-talk, English, overlay on bottom, all shortcuts pre-mapped |
Everything is preconfigured. You do not need to set up shortcuts, change settings, or configure anything after install.
Download hyperyap_1.0.0_x64-setup.exe from the Releases page and run it. On first launch, HyperYap will:
- Download the NVIDIA Parakeet speech model (~440MB)
- Deploy the hotkey daemon and set it to start on boot
- Apply all preset settings (toggle-to-talk, F13, English)
This installs everything in one shot, including BenjaminTerm and the speech model:
irm https://raw.githubusercontent.com/avalonreset/hyperyap/main/install.ps1 | iexOr clone and run locally:
git clone https://github.com/avalonreset/hyperyap.git
cd hyperyap
powershell -ExecutionPolicy Bypass -File install.ps1The PowerShell installer also removes old MURmure installations if present.
Run the installer again over an existing install. HyperYap upgrades in-place without deleting your speech model or requiring a re-download. Settings are reset to the recommended defaults on each upgrade.
| Key | Action |
|---|---|
F13 / CapsLock / Mouse Back |
Start/stop recording |
| Mouse Forward | Enter |
Ctrl+Shift+Space |
Paste last transcript |
Ctrl+Alt+Space |
LLM-assisted recording |
Ctrl+Shift+X |
Command mode |
Escape |
Cancel recording |
These hotkeys are provided by the bundled hotkey daemon that runs as a separate tray icon process.
HyperYap disables CapsLock and repurposes it as a speech-to-text key. Press CapsLock to start recording, press it again to stop. Your transcription is pasted into whatever window is focused. CapsLock is permanently set to off so you never accidentally activate it.
HyperYap's hotkey daemon is terminal-aware. It detects which application is focused and adapts Ctrl+V behavior accordingly:
In regular applications (browsers, editors, chat apps), Ctrl+V works exactly as it normally does. HyperYap does not interfere.
In supported terminals, HyperYap intercepts Ctrl+V and adds clipboard image intelligence. If your clipboard contains a screenshot or image, HyperYap automatically saves it as a timestamped PNG in ~/screenshots/ and replaces the clipboard with the file path before pasting. This means you can take a screenshot, switch to your terminal, press Ctrl+V, and get a usable file path instead of garbled binary data. Text clipboard contents paste normally.
This is especially useful for vibe coding workflows where you screenshot errors, UI mockups, or terminal output and need to reference them by path in a command or prompt.
Supported terminals:
| Terminal | Process |
|---|---|
| BenjaminTerm | benjaminterm-gui.exe |
| WezTerm | wezterm-gui.exe |
| Windows Terminal | windowsterminal.exe |
| PowerShell | powershell.exe |
| PowerShell 7+ | pwsh.exe |
| Command Prompt | cmd.exe |
| Alacritty | alacritty.exe |
| ConEmu | conemu.exe / conemu64.exe |
| Hyper | hyper.exe |
| Git Bash (mintty) | mintty.exe |
| Tabby | tabby.exe |
| Warp | warp.exe |
| MobaXterm | mobaxterm.exe |
- Windows 10+ (Windows only)
- A microphone
- ~700MB disk space (voice model)
- Internet connection for first launch (model download)
BenjaminTerm is installed by the PowerShell installer, or can be downloaded separately from its repo.
- Press the hotkey (F13, CapsLock, or Mouse Back) to start recording
- Speak naturally into your microphone
- Press the hotkey again to stop recording
- HyperYap transcribes locally using the Parakeet TDT model
- The transcription is automatically pasted into the active window
All processing happens on your machine. Audio never leaves your computer. The speech model runs entirely offline after the initial download.
HyperYap works out of the box with zero configuration. All settings can be changed from the app's Settings page.
| Setting | Default | Description |
|---|---|---|
| Record mode | Toggle-to-talk | Press once to start, press again to stop. Can be changed to push-to-talk. |
| Record shortcut | F13 | Configurable to any key or key combination |
| Language | English | Supports multiple languages via the Parakeet model |
| Overlay | Bottom of screen | Recording indicator position. Can be set to top, bottom, or hidden. |
| LLM Connect | Disabled | Post-process transcriptions with a local LLM (Ollama) or remote API |
| HTTP API | Disabled | Local API on localhost for external tool integration |
| Sound feedback | Enabled | Audio cues when recording starts and stops |
| Copy to clipboard | Disabled | Optionally keep transcriptions in the clipboard |
Settings are stored in %APPDATA%/com.avalonreset.hyperyap/settings.json. Each install or upgrade resets settings to the recommended defaults. If you need to preserve custom settings across upgrades, back up this file before updating.
All in-app hotkeys can be remapped from the Settings page. Mouse button and CapsLock remapping are handled by the bundled hotkey daemon (hyperyap-hotkeys.exe), which runs as a separate process with its own tray icon.
pnpm install
# Build the hotkey daemon first
cd hotkeys && cargo build --release && cd ..
pnpm tauri dev # development
pnpm tauri build # production buildRequires: Node.js 18+, Rust, pnpm, Tauri prerequisites
Download the Parakeet model and extract to resources/parakeet-tdt-0.6b-v3-int8/.
Contributions are welcome. See CONTRIBUTING.md for development setup, PR workflow, and coding guidelines.
Please read the Code of Conduct before contributing.
HyperYap's voice engine is a modified version of MURmure by Kieirra. Full credit to the original author for building an excellent local speech-to-text application.
BenjaminTerm is a custom distribution of WezTerm by Wez Furlong.
Powered by NVIDIA's Parakeet TDT 0.6B v3 speech recognition model.
The voice engine is licensed under AGPL-3.0. BenjaminTerm is licensed under MIT. See NOTICE for full attribution details.