Open Source · Python · AI
Voice-Controlled Graphics.
Draw with your voice. A Python system powered by AI speech recognition — making creativity accessible to everyone.
SCROLL
What it is
Built to democratize
creative coding.
Voice → Shape
Speak natural commands. Say "draw a square" or "circle with radius 50" and watch it appear instantly on screen.
AI-Powered Recognition
Google's Speech API processes continuous dictation, converting spoken language into executable graphics commands in real time.
Built for Everyone
Designed to empower physically challenged individuals — removing the keyboard and mouse as barriers to creative expression.
Developer Contributions
What Danish
built & fixed.
01
Voice command handling overhaul — Refactored the core command loop for faster response time and cleaner parsing logic.
Feature
02
Number word conversion fix — Patched the
"ten" → 10 bug that caused all dimension commands to fail silently.Bug Fix
03
Undo, clear & save commands — Implemented three critical workflow actions, giving users full control over their canvas state.
Feature
04
Logging & real-time debug support — Added structured logging across all modules with live console output for easier development.
Dev Tools
05
Terminal help assistant — Built an in-terminal guide that surfaces available commands contextually during a session.
Feature
06
Enhanced error handling & feedback — Users are now clearly informed when speech recognition or command execution fails, with graceful recovery.
UX
Performance
Accuracy over time.
Key Features
Everything you
need to create.
Natural Language Commands
No memorizing syntax. Speak like a human — "draw a big red circle" — and the system handles interpretation.
↗ Continuous Dictation
Undo · Save · Clear
Full canvas control through voice. Mistakes are reversible, progress is saveable, and the canvas resets on command.
↗ Non-destructive Workflow
Real-Time Response
Low-latency pipeline between speech input and Turtle graphics output makes the experience feel immediate and fluid.
↗ <500ms latency
Educational by Design
Teaches shape logic, coordinate systems, and basic programming concepts without ever touching a keyboard.
↗ STEM Learning Tool
Impact
Who it helps.
Classrooms
Reduces onboarding time for visual programming. Students learn shape logic and spatial reasoning faster through hands-on, voice-driven interaction.
Accessibility
Removes the physical input barrier for users with motor disabilities. Code and create graphics with nothing but your voice.
E-Learning Platforms
Embeddable in online STEM courses as an interactive drawing module — no installation, no friction for learners.
STEM for All
Lowers the cognitive barrier to entry for programming. If you can speak, you can code graphics — language-agnostic by design.
Roadmap
What's coming next.
Phase 1 · In Progress
3D Shape Drawing via OpenGL
Extend the graphics engine from 2D Turtle to full 3D rendering using PyOpenGL, enabling spatial voice commands like "draw a cube" or "rotate the sphere."
Phase 2 · Planned
Speech-to-Code Expansion
Bridge the gap between VCG and full Python — students can speak programs aloud, and VCG will transcribe them into executable code, teaching syntax without typing.
Phase 3 · Research
Touch & Gesture Recognition
Integrate sensor-based input to combine gesture control with voice — a multimodal creative environment for users of all abilities.
Phase 4 · Vision
Web-Based VCG with WebGL
Port the entire system to the browser using JavaScript, WebSpeech API, and WebGL — no installation required, accessible globally from any device.