Skip to content

ARCHITECTURE.md

Latest commit

 

History

History
34 lines (28 loc) · 1.44 KB

ARCHITECTURE.md

File metadata and controls

34 lines (28 loc) · 1.44 KB
Raw

Architecture — CortexLearn

High-Level Design (HLD)

CortexLearn listens to a learner over WebRTC, transcribes with Whisper, and adapts the learning path using a knowledge graph of concepts and mastery — a spoken, personalized tutor.

%%{init: {'theme':'base','themeVariables':{'primaryColor':'#ffffff','lineColor':'#2563eb','mainBkg':'#ffffff'}}}%%
graph LR
    A([Voice])
    B([Whisper STT])
    C([Adaptive Engine + KG])
    D([Personalized Path])
    A --> B
    B --> C
    C --> D
    style A fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#1e40af
    style B fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#1e40af
    style C fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#1e40af
    style D fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#1e40af
Loading

Flow: Voice → Whisper STT → Adaptive Engine + KG → Personalized Path

Low-Level Design (LLD)

  • Components: WebRTC, Whisper, Knowledge Graph
  • Interfaces / contracts: to be finalized during implementation.
  • Data model: to be defined per component.

Decision Log

  • Why this stack: WebRTC — real-time media transport; Whisper — speech-to-text model; Knowledge Graph — entity & relationship store.
  • ** constraint:** run logic/state/UI locally; offload heavy reasoning to cloud APIs; target modest hardware.

Concept Deep Dive

Modeling learner mastery well enough to adapt content in real time from spoken interaction.