Screamer

Fast Windows dictation that types wherever your cursor is.

Press a hotkey, speak, release - Screamer records your voice, sends it to a speech-to-text provider, optionally cleans up the result with an LLM, and types the final text into the active window.

No browser tab. No copy-paste. Just talk and keep moving.

Install

Download the latest release from the Releases page.

Grab the versioned Windows zip:

Screamer-vX.Y.Z-windows-x64.zip

Then:

1. Extract the zip.
2. Open the Screamer folder.
3. Run Screamer.exe.
4. Configure your provider in Settings.

That's it.

What Screamer does

• Global hotkey dictation - speak from anywhere on Windows.
• Hold-to-talk or toggle mode - choose how recording should behave.
• Types into the focused app - works in editors, browsers, chats, notes, docs, and more.
• OpenAI-compatible speech-to-text - use OpenAI, Groq, or another compatible /audio/transcriptions endpoint.
• Optional AI cleanup - fix punctuation, grammar, spelling, and capitalization after transcription.
• Fallback providers - configure backup STT and LLM providers if the primary one fails.
• On-screen recording indicator - a small pulsing pill appears at the bottom-center of the screen while recording and processing.
• System tray app - enable/disable, change hotkey, toggle rewrite, open settings, or exit from the tray.
• Microphone selection - pick your input device and calibrate silence detection.
• Post-type key - optionally press Enter, Tab, Space, or Backspace after typing.
• Windows startup support - launch Screamer automatically when you log in.
• Secure API key storage - API keys are stored locally with Windows DPAPI.

How it works

Hotkey -> Record audio -> Transcribe -> Optional cleanup -> Type into active window

Screamer records 16 kHz mono WAV audio, sends it to your configured STT provider, optionally runs the text through an LLM cleanup step, then injects the final text with Windows SendInput.

Setup

On first launch, open Settings from the tray icon.

You need at least one speech-to-text provider.

Example OpenAI-compatible STT config:

Base URL: https://api.openai.com/v1
Model: whisper-1
API key: your_api_key

For Groq, prefer:

Base URL: https://api.groq.com/openai/v1
Model: whisper-large-v3-turbo
Language: en

If accuracy matters more than speed, use whisper-large-v3 instead.

For another OpenAI-compatible provider, use its base URL and model name.

The LLM rewrite step is optional. Leave it off if you want raw transcription.

Settings

General

• Recording mode: Hold to talk or Toggle
• Hotkey selection
• Post-type key
• Start with Windows

STT

• Primary speech-to-text provider
• Optional fallback STT provider
• Optional transcription language
• Custom headers

LLM

• Optional AI rewrite
• Primary LLM provider
• Optional fallback LLM provider
• Editable system prompt
• Custom headers

Audio

• Input device selection
• Silence threshold calibration

Hotkeys

Quick-pick presets:

Ctrl+Alt+Space
Ctrl+Shift+Space
Ctrl+Alt+D
Ctrl+Alt+S
Ctrl+Alt+V
Scroll Lock
Pause

Default: Ctrl+Alt+Space

Or set a custom hotkey: in Settings, click Record and press any key combination, a function key, or a mouse side/middle button. Bare everyday keys need a modifier (Ctrl/Alt/Shift); function keys, lock/pause keys, and mouse side/middle buttons may be bound on their own. The matched trigger is swallowed so it won't reach the app underneath.

For developers

Run from source:

pip install -r requirements.txt
python -m src.main

Run tests:

python -m unittest discover -s tests -v

Build dependencies:

pip install -r requirements-build.txt

Build the Windows executable:

python -m PyInstaller --noconfirm --clean screamer.spec

Releases

Pushing a version tag builds and publishes a Windows release automatically:

git tag v1.0.0
git push origin v1.0.0

The release workflow:

1. Installs dependencies.
2. Runs the test suite.
3. Builds Screamer with PyInstaller.
4. Packages the app as:

Screamer-v1.0.0-windows-x64.zip
Screamer-v1.0.0-windows-x64.zip.sha256

Hyphenated tags like v1.0.0-rc1 are published as pre-releases.

Platform

Screamer is built for Windows.

It depends on Windows-specific features including:

• global hotkeys (keyboard or mouse) via low-level hooks (WH_KEYBOARD_LL/WH_MOUSE_LL)
• text injection via SendInput
• tray integration
• DPAPI key storage
• startup registration through the current user Run key

License

MIT