YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

𐀀 NUMID.AI

Ancient languages, future minds.

A futuristic AI-powered translation platform that bridges 5,000 years of human writing β€” from Sumerian cuneiform to Phoenician abjad, Egyptian hieroglyphs to Old Norse runes β€” powered by Anthropic Claude.


✨ What It Does

Numid.ai translates modern text into 17 ancient and classical writing systems, rendering authentic Unicode scripts with:

  • Authentic script rendering β€” real Unicode characters for each writing system
  • Phonetic transliteration β€” IPA and scholarly romanization
  • Word-by-word breakdown β€” sign-by-sign analysis
  • Cultural & historical context β€” why certain words/signs were chosen
  • Pronunciation guide β€” how ancient peoples would have sounded
  • Confidence ratings β€” honest about reconstructed languages
  • Cross-language comparison β€” same phrase in 6 scripts simultaneously
  • Historical timeline β€” visual journey through 5,000 years of writing

πŸ—Ώ Supported Ancient Languages (17)

Language Era Script Direction
Phoenician 1050–150 BCE Phoenician Abjad RTL
Sumerian 3500–2000 BCE Cuneiform LTR
Akkadian 2800–100 BCE Cuneiform LTR
Ancient Egyptian 3200 BCE–4th c CE Hieroglyphics RTL
Proto-Sinaitic 1900–1500 BCE Pictographic RTL
Ugaritic 1400–1185 BCE Cuneiform Alphabet LTR
Imperial Aramaic 800 BCE–600 CE Aramaic Abjad RTL
Old Persian 600–300 BCE Cuneiform LTR
Linear B 1450–1200 BCE Syllabary LTR
Gothic 350–800 CE Gothic Alphabet LTR
Etruscan 700–100 BCE Old Italic RTL
Classic Maya 250–900 CE Maya Glyphs LTR
Vedic Sanskrit 1500–200 BCE Brahmi LTR
Classical Latin 75 BCE–600 CE Roman LTR
Ancient Greek 800–300 BCE Greek LTR
Old Norse 800–1300 CE Runes LTR
Coptic 200–17th c CE Coptic LTR

πŸ—‚οΈ Repository Structure

numid-ai/
β”‚
β”œβ”€β”€ backend/                        # Python FastAPI server
β”‚   β”œβ”€β”€ main.py                     # App entry point, CORS, routes mounting
β”‚   β”œβ”€β”€ requirements.txt            # Python dependencies
β”‚   β”œβ”€β”€ __init__.py
β”‚   β”‚
β”‚   β”œβ”€β”€ models/
β”‚   β”‚   β”œβ”€β”€ __init__.py
β”‚   β”‚   └── languages.py            # Full language catalogue (17 languages)
β”‚   β”‚
β”‚   β”œβ”€β”€ routers/
β”‚   β”‚   β”œβ”€β”€ __init__.py
β”‚   β”‚   └── translate.py            # All API endpoints
β”‚   β”‚
β”‚   β”œβ”€β”€ services/
β”‚   β”‚   β”œβ”€β”€ __init__.py
β”‚   β”‚   β”œβ”€β”€ translator.py           # Anthropic Claude translation engine
β”‚   β”‚   └── audio.py                # TTS pronunciation generation
β”‚   β”‚
β”‚   └── utils/
β”‚       └── __init__.py
β”‚
β”œβ”€β”€ frontend/                       # React application
β”‚   β”œβ”€β”€ package.json
β”‚   β”œβ”€β”€ public/
β”‚   β”‚   └── index.html
β”‚   └── src/
β”‚       β”œβ”€β”€ index.js                # React entry point
β”‚       β”œβ”€β”€ App.jsx                 # Root component, nav, star field
β”‚       β”‚
β”‚       β”œβ”€β”€ pages/
β”‚       β”‚   β”œβ”€β”€ TranslatorPage.jsx  # Main translation UI
β”‚       β”‚   β”œβ”€β”€ TimelinePage.jsx    # Historical language timeline
β”‚       β”‚   └── ComparePage.jsx     # Multi-language comparison
β”‚       β”‚
β”‚       β”œβ”€β”€ components/
β”‚       β”‚   β”œβ”€β”€ LanguageCard.jsx    # Individual language selector card
β”‚       β”‚   └── TranslationResult.jsx # Full translation display
β”‚       β”‚
β”‚       β”œβ”€β”€ hooks/
β”‚       β”‚   └── useTranslation.js   # API hooks (languages, translate, compare)
β”‚       β”‚
β”‚       └── styles/
β”‚           └── globals.css         # Design system, variables, animations
β”‚
β”œβ”€β”€ docs/                           # Documentation
β”‚
β”œβ”€β”€ .env                            # Your config (not committed)
β”œβ”€β”€ .env.example                    # Template
β”œβ”€β”€ .gitignore
└── README.md

πŸš€ Quick Start

Prerequisites

1. Setup

unzip numid-ai.zip
cd numid-ai
cp .env.example .env

Edit .env and set your ANTHROPIC_API_KEY.

2. Backend

cd backend
python -m venv venv
source venv/bin/activate       # Windows: venv\Scripts\activate
pip install -r requirements.txt
cd ..
python -m backend.main
# β†’ API running at http://localhost:8000
# β†’ Docs at http://localhost:8000/docs

3. Frontend

cd frontend
npm install
npm start
# β†’ App at http://localhost:3000

πŸ”Œ API Reference

GET /api/v1/languages

Returns all 17 languages with full metadata.

POST /api/v1/translate

{
  "text": "The stars are eternal",
  "target_language_id": "phoenician",
  "source_language": "auto",
  "style": "formal"
}

style options: formal | poetic | conversational | inscription

POST /api/v1/compare

{
  "text": "I am the king",
  "language_ids": ["sumerian", "ancient_egyptian", "phoenician"]
}

POST /api/v1/pronunciation

{
  "script_text": "𐀁𐀏𐀋",
  "language_id": "phoenician"
}

GET /api/v1/timeline

Languages grouped by historical era.


βš™οΈ Environment Variables

Variable Required Description
ANTHROPIC_API_KEY βœ… Powers all translations via Claude
HOST no Backend host (default 0.0.0.0)
PORT no Backend port (default 8000)
DEBUG no Hot reload in development
TTS_PROVIDER no offline or google for audio
GOOGLE_TTS_API_KEY no Enhanced audio pronunciation
MAX_TOKENS_PER_REQUEST no Claude token limit (default 4096)

🧠 How the AI Translation Works

  1. System prompt instructs Claude as an expert in all 17 ancient language systems
  2. Claude translates the modern text, choosing appropriate signs/letters
  3. Returns structured JSON: script rendering, IPA, word breakdown, cultural notes
  4. For extinct/reconstructed languages, Claude is honest about confidence level
  5. Responses are cached (1 hour) to minimize API calls
  6. Parallel translation available for cross-language comparison

πŸ›οΈ Philosophy

"Every alphabet you know descends from Phoenician merchants scratching consonants onto clay in 1050 BCE. Every number you write traces to Sumerian accountants in Ur. We don't start from scratch β€” we stand on 5,000 years of human ingenuity."

β€” Numid.ai


πŸ“ License

MIT β€” free to use, fork, and build upon. If you build something cool, share it.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support