SurvivalRussian

AI-assisted multimodal Russian learning system focused on real-world language production.

🧠 Overview

SurvivalRussian is a full-stack language learning platform designed around active recall and output-based learning, rather than passive recognition.

The system integrates typing, speaking, listening, and contextual understanding into a single feedback loop to accelerate practical language acquisition.

⚙️ Core Features ⌨️ Cyrillic Typing Engine Learn Russian through direct keyboard input (RU layout) Built-in typing trainer for users unfamiliar with Cyrillic Reinforces spelling, muscle memory, and recall simultaneously 🔊 Speech + Pronunciation Feedback Text-to-speech playback for all words and phrases User voice recording and sampling Audio is sent to an AI API for pronunciation analysis Returns concise feedback on: mispronounced sounds stress issues phonetic inaccuracies 🧠 Contextual Vocabulary System

Each word includes:

Natural meaning + literal meaning Grammatical usage Synonyms & antonyms Visual association (image-based reinforcement) 📖 Grammar Breakdown Engine Sentences include AI-generated grammatical explanations Short-form (1–2 paragraph) breakdowns Focused on practical understanding, not academic overload 🤖 AI-Generated Lessons Dynamically generated typing and practice exercises Adapts to user interaction and learning flow 💬 Collaborative Learning Live chat system for shared learning and discussion Enables real-time interaction between users 🧩 (In Progress) Morphology Tools Interactive exploration of: prefixes roots suffixes Planned “slot-machine” style interface for constructing words Case manipulation tools for understanding inflection patterns 🎯 Learning Philosophy

SurvivalRussian is built around a few core principles:

Output over input → typing and speaking > recognition Recall over review → memory is built through effort Context over memorization → meaning is learned in use Mobility over static study → designed for use during real-life activity

The system is heavily inspired by audio-first learning approaches (e.g., Pimsleur), while extending them with interactive and AI-driven feedback loops.

🛠️ Tech Stack Next.js (App Router) JavaScript / React Text-to-Speech APIs Speech analysis via AI API integration Real-time communication (chat)

(Expand this if you want to flex more specifics)

🚧 Status

Proof of concept with core systems implemented. Currently expanding:

sentence database morphology tools lesson generation depth 🚀 Running Locally npm install npm run dev

WSL users:

npm run dev:wsl

Then open: http://localhost:3000