Back to browse
VoiceDraw – Talk system design out loud, the diagrams draw themselves

VoiceDraw – Talk system design out loud, the diagrams draw themselves

by ajaypanthagani·Jun 16, 2026·47 points·19 comments

AI Analysis

●●SolidSlickSolve My Problem

Real-time voice-to-diagram with reasoning capture—Eraser and Miro AI don't do this live.

Strengths
  • Renders diagrams in real-time while you speak, not after you finish describing
  • Captures open questions and tradeoffs alongside the visual diagram automatically
  • Clean UI with Google sign-in and guided tour for immediate usability
Weaknesses
  • AI diagram space heating up fast with Eraser, Miro AI, Lucidchart plugins
  • Voice-to-diagram accuracy will live or die on edge cases in complex architectures
Category
Target Audience

Engineers doing system design discussions, interviews, or team architecture sessions

Similar To

Eraser · Miro AI · Lucidchart

Post Description

I was frustated by having to draw system design diagrams by hand when discussing it with my team mates or in an interview. So I thought "Wouldn't it be great if someone draws it for me, while I think out loud?".

That is when I came up with VoiceDraw. You can just think out loud or discuss your system architecture with a friend/interviewer, the diagrams are automatically drawn along with your reasoning, open questions and tradeoffs beautifully written on to the side.

Demo Video: https://youtu.be/36PgHKSuccE

Similar Projects

Design●●Solid

Napkin – desktop app for quick diagrams, with MCP support

Napkin keeps everything local while giving you an embeddable MCP endpoint so AI agents can draw and edit diagrams at http://127.0.0.1:21420/mcp. It combines Excalidraw-style hand-drawn tooling (rough.js, connectors, grid snapping, PNG/SVG export) with a Tauri/Rust desktop shell and delta-compressed version snapshots — a neat, concrete take on AI-assisted diagramming, though its impact hinges on MCP adoption and the surrounding agent ecosystem.

Niche GemWizardry
ipcrm
304mo ago