Ideate, sketch, and edit diagrams in real time on a voice-first canvas. Speak naturally to build system architectures, flowcharts, and turn ideas into visuals.
10x faster than Claude code with Excalidraw skill
No manual dragging. No long prompts. No Excalidraw skill setup. Just talk to the canvas and refine as you go.
The .excalidraw export is the detail that sells it for me — means the output stays real editable elements, not a flattened image I'd have to redraw. Once I export a diagram and tidy it up by hand in Excalidraw, can I bring that file back into BUD and keep refining it by voice, or is the voice editing one-way out? Trying to figure out if BUD is the whole loop for me or just the fast first-draft step.
Voice-driven diagramming sounds genuinely novel, love that you're tackling this. One thing I'd want as a user: the ability to easily undo or branch a voice command when the diagram doesn't render quite right, maybe a quick "scratch that" command that rewinds the last addition without losing the whole flow.
the voice-first approach to diagramming feels really natural here, kind of like you're thinking out loud and the canvas just keeps up. love how it stays out of the way until you actually need it.
Sketched out a basic system architecture by just talking through it, way faster than dragging boxes around. Voice commands felt surprisingly natural once I got used to phrasing structure.
Sketching a system diagram by just talking to it feels weirdly natural, and seeing the boxes appear as I described relationships was surprisingly satisfying.
Voice-to-diagram sounds gimmicky but it actually kept up when I described a basic microservices flow. Would love to see how it handles edits mid-sketch.
The voice-to-diagram flow here feels surprisingly natural, especially how it handles rough conceptual ideas and turns them into clean shapes without endless manual cleanup. Really thoughtful UX choice.
Spoke through a basic system architecture and it actually mapped out the boxes and arrows as I went, which felt a bit like magic. The voice recognition got most of my jargon right on the first try.
Honestly the voice-to-diagram thing is way smoother than I expected, basically just talked through a system architecture and it laid out the boxes and arrows as I went. Kind of feels like sketching with a coworker who never gets tired of your rambling.
Spoke through a quick system architecture and was surprised how it actually understood relationships between boxes without me being super precise. Editing with voice feels a little weird at first but it grows on you fast.
Voice-to-diagram actually worked on my rambling description of a microservices setup. Sketching flowcharts by talking feels weirdly natural, though I wish it understood when I said "connect those two" a bit better.
Spoke through a basic cloud setup and watched it map out boxes and arrows without me touching the keyboard. The voice parsing is surprisingly accurate for technical terms like load balancer and S3 bucket.
Tried sketching a quick system architecture on BUD just by talking through it, and the diagram actually kept up with my rambling. Latency felt impressively low for a voice-driven canvas.
Finally something for folks who think out loud, this actually worked when i described a simple API flow and it drew the boxes and arrows correctly. kind of wild how natural it feels to just talk and watch a diagram appear.
The voice-first canvas is such a smart move for diagramming, especially when you're mid-thought and don't want to break flow clicking shapes. Curious how it handles ambiguity when you rephrase something mid-sketch.
The voice-first canvas concept is genuinely refreshing for diagramming, which has felt stale for years. Excited to see how the natural language parsing handles ambiguous ideas like "make it connect better" or "show the flow back to auth."
The voice-first canvas feels surprisingly natural for sketching out system diagrams hands-free. Liked that it kept up with my rambling about microservices without losing the structure.
the idea of building out diagrams by just speaking feels so natural honestly, like the canvas stays out of your way while you talk through ideas
the voice-to-diagram thing actually works pretty well, like i just described a basic flow and it popped up roughly what i meant. honestly the realtime editing on the canvas is slick
About BUD on Product Hunt
“Voice-first canvas for sketchnoting and whiteboarding”
BUD launched on Product Hunt on July 21st, 2026 and earned 111 upvotes and 43 comments, placing #24 on the daily leaderboard. Ideate, sketch, and edit diagrams in real time on a voice-first canvas. Speak naturally to build system architectures, flowcharts, and turn ideas into visuals.
BUD was featured in Design Tools (261.4k followers), Productivity (656.6k followers) and Sketch (18.8k followers) on Product Hunt. Together, these topics include over 188.6k products, making this a competitive space to launch in.
Who hunted BUD?
BUD was hunted by Aryan Bhargav. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how BUD stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Aryan Bhargav