Audio Retrieval SRS

Audio-first spaced repetition system for language learning with hands-free practice sessions.

Architecture

Monorepo structure:

• packages/core - Shared TypeScript domain models and scheduling engine
• packages/storage - Storage adapters (IndexedDB for web, SQLite for mobile)
• apps/web - Next.js web application
• apps/mobile - Expo React Native iOS application

Setup

Prerequisites

• Node.js 18+
• pnpm 8+
• FFmpeg (for audio segmentation) - Download FFmpeg and ensure it's in your PATH
• OpenAI API key (for Whisper transcription)

Installation

pnpm install

Environment Setup

Create a .env.local file in the apps/web directory:

OPENAI_API_KEY=sk-your-openai-api-key-here

Note: The OpenAI API key is required for the Whisper transcription feature. Get your API key from OpenAI.

Development

Web App

pnpm dev:web

Mobile App

pnpm dev:mobile

Build

pnpm build:core
pnpm build:web

Testing

pnpm test:core

Features

• Audio-first practice sessions
• Leitner-like spaced repetition scheduling
• Hands-free operation (speech recognition for ratings)
• Background audio support on iOS
• Local storage with export/import
• Session persistence and recovery
• OpenAI Whisper Integration: Upload audio files to automatically transcribe and translate into sentence pairs
• Multi-language Support: Study multiple languages with separate data storage per language
• Time-aligned Audio Segments: Extract individual sentence audio clips from longer recordings
• Duplicate Detection: Fuzzy matching to identify similar sentences before adding
• Sentence Validation: Automatic filtering to ensure only complete sentences are saved

License

MIT