Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.ajayji.com/llms.txt

Use this file to discover all available pages before exploring further.

The Scribe is designed for mission-critical voice capture. It’s not a real-time conversationalist; it’s a professional stenographer that never loses a word, built on a resilient, offline-first architecture. How it Works: The Three-Phase Workflow
  1. Capture (iOS): Record long-form audio on your iPhone or iPad. The app immediately saves the audio to a local database on your device, ensuring that even if the app is closed or your device loses power, your recording is safe. For immediate feedback, you’ll see a preliminary “ghost text” transcription.
  2. Sync: When you’re back on the same local network as your Mac, the app will automatically and seamlessly sync the audio and its metadata to your host machine.
  3. Process (macOS): Your Mac, with its greater processing power, takes over. It performs a high-accuracy transcription and can then run persona-based AI models to automatically format, summarize, or analyze the text into a polished final document.
Direct Dictation on macOS For users who prefer to work directly on their desktop, aj also provides a direct, local Scribe feature on macOS. This allows you to leverage the same powerful transcription and processing capabilities without needing a separate mobile device. The app also includes a rich UI for managing your transcription models, allowing you to activate, deactivate, and perform live tests to find the best model for your needs. Key Benefits:
  • Zero Data Loss: The offline-first, database-backed design guarantees that your important thoughts are saved locally and securely.
  • Resilient and Reliable: The asynchronous sync and processing model means that network interruptions or app closures won’t cause you to lose your work.
  • Private and Secure: Your voice data is captured and stored on your devices, and the entire workflow can operate on your local network, ensuring your data stays private.
  • You’re in Control: From the initial capture to the final processing, you have control over the entire workflow, including which models are used to transcribe and summarize your dictations.