Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/jtapieromalambo-ctrl/Signia/llms.txt

Use this file to discover all available pages before exploring further.

Signia is a Django-based web application that bridges spoken and written Spanish with Colombian Sign Language (LSC). It offers two complementary translation modes: text and audio to sign language videos, and real-time webcam sign recognition to text — making communication more accessible for deaf and mute users.

Introduction

Learn what Signia is, how it works, and what it can do.

Quickstart

Run Signia locally in minutes with a step-by-step guide.

Text to Signs

How Spanish text and audio are translated into LSC videos.

Sign Recognition

Real-time webcam recognition powered by MediaPipe and RandomForest.

How It Works

Signia combines several technologies to enable bidirectional communication between Spanish and LSC:
1

User inputs text or audio

A user types Spanish text or records a voice message in the translator interface at /traductor/.
2

LSC grammar conversion

The input is sent through lsc_grammar.py, which uses the Groq API to reorder words into the correct SOV (Subject–Object–Verb) LSC grammatical structure.
3

Video lookup and playback

Each LSC token is matched to a pre-recorded sign video stored in the database and played back in sequence.
4

Webcam sign recognition (reverse flow)

In the recognition module at /reconocimientos/camara/, the webcam captures hand movements, MediaPipe extracts landmarks, and a RandomForest classifier identifies the sign in real time.

Explore the Docs

Installation

Full setup guide with environment variables, database, and static files.

Configuration

All required and optional environment variables explained.

Deploy on Railway

Deploy Signia to production on Railway with nixpacks.

API Reference

REST endpoints for recognition, prediction, and admin operations.

Build docs developers (and LLMs) love