How yeheskieltame/linggo-english-hub works
Project type and purpose
• This is a single-page web application (SPA) built with Vite, TypeScript, and React.
• Purpose: a language-learning platform (Lovable) that presents CEFR-based lessons, quizzes, reading/listening activities, conversation practice, and AI-driven feedback via the Groq LLM API.
• It uses Supabase for authentication and persistence and the browser’s Web Speech APIs for speech recognition and text-to-speech.
File structure
• Top-level:
• package.json, Vite/Tailwind/PostCSS configs → build & bundling
• supabase/ → local Supabase configuration & migration scripts
• public/ → static assets
• src/ → application code
• Under src/:
• pages/ → React Router pages (Dashboard, Lessons, Auth, Profile…)
• components/ → UI building blocks (shadcn-ui wrappers, page fragments, shared UI library)
• services/ → network and business-logic services (supabase client, llmService, conversationService, speechRecognition, textToSpeech, levelAssessmentService)
• integrations/supabase → supabase-js client initialization and types
• providers/AuthProvider.tsx → React Context for user auth state
• data/ → lesson content and mock data loaded at build time
• hooks/ → custom React hooks (useMobile, useToast)
README for additional insights
• Tech stack: Vite, React, TypeScript, Tailwind CSS, shadcn-ui, Groq LLM API, Supabase.
• No custom Node-based backend; all “backend” functionality is offloaded to Supabase and external HTTP APIs.
• Environment variables: VITE_GROQ_API_KEY for AI calls.
• Deployment via Lovable (static hosting + functions) or any generic static host with environment support.