Design and Implementation of AI Planner: An Intelligent Daily Planning and Habit Tracking Framework with Speech-to-Intent NLP and Affective Analytics

Traditional personal organization software and habit tracking applications require extensive manual interaction, rigid form filling, and offer zero contextual feedback regarding a user's psychological or physical capacity. These interaction barriers exacerbate the intention-behavior gap, resulting in high abandonment rates. This paper presents the design, architecture, and implementation of AI Planner, an intelligent end-to-end productivity ecosystem designed to eliminate data entry friction and provide data-driven behavioral insights. The system integrates: (1) a natural language understanding (NLP) engine for zero-friction task capture with automatic priority and category inference; (2) a hands-free speech-to-intent voice processing pipeline that transcribes and categorizes user requests into tasks, recurring habits, or long-term goals; (3) an intelligent hierarchical goal decomposition engine that automatically subdivides complex objectives into actionable, ordered milestones; and (4) an empirical statistical correlation engine that analyzes longitudinal relationships between daily mood levels, self-reported energy scores (1-5), and habit adherence over rolling temporal windows. The platform is engineered across a 3-tier decoupled architecture comprising a cross-platform Flutter client, an asynchronous FastAPI application engine, and a fully normalized MySQL relational database. We demonstrate the functional validity, operational workflows, relational integrity, and practical benefits of AI Planner as a scalable solution for modern self-regulation and personal informatics.