Case study
Open sourceVoice to Text + Instant Translate
Windows tray tool: double-tap ALT to dictate anywhere via Whisper, or select text and double-tap CTRL for an instant floating translation.
Facts
- Status
- Open source
FIG. 001 / ARCHITECTURE
How the system connects
Live model
Problem
Typing long messages or switching between Turkish and English content on Windows breaks flow — there was no fast, keyboard-only way to dictate text into any app or get an instant translation without copying text into a separate translator.
System
A background Python app listens for global hotkeys: double-tapping ALT starts/stops a recording that is sent to the Whisper API and auto-pasted into whatever field has focus (with smart window-memory so it returns to the right app), while double-tapping CTRL on selected text opens a floating popup with a Turkish translation, and CTRL+ALT replaces selected text in place with an English translation. A public macOS twin repeats the same double-tap-to-record idea natively in Swift.
FIG. 002 / HOW IT WORKS
How it works
A minimal, always-on Windows productivity tool built to remove the friction of dictating and translating text mid-workflow. Rather than opening a separate app, the entire interaction is two keyboard taps: speak and it's pasted, or select text and it's translated in a popup next to the cursor. A companion macOS app carries the same double-tap-to-dictate idea to Mac using native Swift, with Whisper transcription and optional GPT-4o-mini cleanup. Both ship as public, documented open-source tools with installers.
Stack
What it runs on
Related systems
Voice Typer
Internal toolA push-to-talk Windows dictation and translation tool with a fully local speech pipeline option (faster-whisper + Ollama) alongside cloud APIs.
UBA-BRAIN
Internal toolA shared markdown knowledge base and rule system that lets any AI CLI on any machine run the same client work with the same rules, memory and skills.
AI Product Visuals Pipeline
Internal toolA skill-driven pipeline that turns a client's chat request into an on-brand AI product photo, picking the right account, model and reference automatically.
Want a system like this for your team?
We start by mapping one workflow on a 30-minute call.