📊
Voice → visual diagram
Mind maps, flowcharts, and sequence diagrams generated from a spoken thought. The output is something you can look at, not just read.
Features
Voice input, on-device classification, diagram rendering, and shareable output — a combination that does not exist together anywhere else.
🎙️
Screen-off capture
A volume button or voice trigger starts recording with the phone locked. No unlock, no app navigation, no broken walk — haptic confirmation when capture completes.
🧠
On-device classification
Five thought types classified locally via Core ML. No voice audio leaves the device. A confidence score is shown, and one-tap correction trains the model.
🔗
Shareable artefact links
Every diagram gets a share URL at share.thoughtsift.app. Async sharing is built in — a voice note becomes a link a colleague can react to without installing anything.
📤
Direct PKM export
Notion direct-write (Pro) pushes a database page with the Type set automatically. Obsidian via URI scheme. Zero manual routing.
🔒
Export-and-evaporate
After export, one tap deletes both the transcript and the cached audio from the device.