All Projects

Dictaro

Planning

Speak messy. Get clean.

Privacy-first AI voice dictation that transforms natural speech into polished, publication-ready text — without sending a word to the cloud.

Productivity Changelog Source
Dictaro landing: speak messy, get clean — a cassette-tape pipeline with transcribe, filler-strip, tone, and dictionary stages

Key Features

  • Local Whisper processing — zero cloud dependency
  • AI cleanup removes ums, stutters, and fixes grammar
  • Works across every app and platform
  • Custom vocabulary and domain-specific jargon support
  • Accessibility-first design for RSI and disability users

Built For

Knowledge workers, people with RSI/disabilities, ADHD individuals, and displaced Dragon users

Vision

Voice dictation is broken. Built-in dictation quality is declining, Dragon costs $699, and Wispr Flow had a privacy scandal. Dictaro bridges the gap with local Whisper processing and intelligent AI cleanup — understanding what you meant to say, not just what you said.

The Problem

Apple's dictation quality thread has 509 "Me Too" responses. The tools that work are either prohibitively expensive (Dragon at $699), privacy nightmares (cloud-processed voice data), platform-locked (Superwhisper is macOS only), or dumb (Voice In has 600K users but no AI intelligence).

Key Differentiators

  • Privacy-first architecture: All processing happens on-device using Whisper. Zero data leaves your machine, ever
  • Intelligent cleanup: Not just transcription — AI understands context, removes filler, fixes grammar, and formats for the target medium
  • Cross-platform: macOS, Windows, Linux, iOS, Android — one subscription, all platforms