Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/GustavoNightmare/InformacionMuseo/llms.txt

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

BioScan Museo turns a museum’s Birds Hall into an interactive digital experience. Visitors scan QR codes to pull up rich species profiles, ask questions to an AI guide powered by a local LLM with Retrieval-Augmented Generation (RAG), and hear narrated descriptions through a text-to-speech sidecar. The platform is self-hosted and runs on low-power hardware such as an Orange Pi 4 Pro using Docker Compose.

Introduction

Understand the architecture, components, and how everything fits together.

Quickstart

Get BioScan Museo running locally in minutes with Docker Compose.

Configuration

Configure environment variables, the LLM backend, and TTS voice settings.

API Reference

Explore the chat, species, metrics, and TTS service endpoints.

What BioScan Museo Does

BioScan Museo is a full-stack platform with two services — a Flask web app and a FastAPI TTS sidecar — that together provide:

QR Code Scanning

Each exhibit has a unique QR code. Scanning it logs the visit and opens the species detail page with image, taxonomy, and habitat info.

RAG Chatbot

Visitors chat with an AI guide that answers questions using Ollama LLM plus ChromaDB vector search over museum documents.

Text-to-Speech

The TTS sidecar pre-generates MP3 narrations for every species and serves them on demand via HTTP or real-time WebSocket.

Admin Panel

Admins manage species, upload museum documents, customize QR styles, and view scan metrics and audit logs.

Get Up and Running

1

Clone the repository

git clone https://github.com/GustavoNightmare/InformacionMuseo.git
cd InformacionMuseo
2

Configure environment variables

Copy the example env file and fill in your credentials:
cp .env.example .env
At minimum, set SECRET_KEY, ADMIN_USER, ADMIN_PASS, MUSEO_TTS_SHARED_KEY, and your Ollama model settings.
3

Start all services

docker compose up -d --build
The stack brings up museo-app (Flask), ollama (LLM), servertts (FastAPI TTS), and two ngrok tunnels for public HTTPS access.
4

Open the app

Navigate to http://localhost:5000. Log in with your admin credentials and use the CLI command flask --app app.py seed to populate a sample species.
The first run downloads Ollama models, which can take several minutes depending on your connection and hardware. See the Deployment guide for Orange Pi 4 Pro-specific tips.

Build docs developers (and LLMs) love