Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

minasamy01/Kemet-AI-Voice-tourism-Assistant

Domaine:

natural language processing

Type de record:

software
Créateur:
min
Hôte:
🏛️ KEMET is an AI-powered smart tourism platform for Egypt. It features a Voice-to-Booking AI agent, real-time translation, and an intelligent monument identification system built to provide a seamless travel experience. # 🏺 KEMET: Voice-to-Booking AI Assistantِ ### *Intelligent Conversational Agent for Smart Egyptian Tourism* --- ## 📖 Overview **KEMET** (كيميت) is a sophisticated **Voice-First AI Agent** designed to modernize the tourism experience in Egypt. Built with **Cohere's Command-R+** and **OpenAI's Whisper**, the system acts as a digital concierge that understands natural voice commands to recommend hotels, provide historical context, and automate the entire booking process from voice to email voucher. ### 🎙️ The Voice-First Revolution Unlike traditional booking apps, Kemet focuses on **Natural Human Interaction**: * **Zero-UI Dependency**: Users can complete a full booking cycle just by speaking. * **Context-Aware Reasoning**: The system doesn't just search; it "reasons" why a specific hotel fits the user's spoken request. --- github.com ## 🏗️ Project Structure ```bash kemet/ # Root Directory ├── 📂 venv/ # Virtual Environment ├── 🐍 app.py # Main Streamlit UI & Session Management ├── 🐍 logic.py # Core AI Pipeline (Whisper & LLM Logic) ├── 📄 hotels.json # Structured Database of Egyptian Tourism ├── 📄 .env # Secret Configuration (API Keys & SMTP) ├── 📄 .gitignore # Git Exclusion Rules ├── 📄 README.md # Professional Documentation └── 📦 requirements.txt # Project Dependencies Manifest ``` --- ## 🧠 Advanced AI Pipeline The system ensures a seamless user journey through a multi-stage intelligent workflow: 1. **Acoustic Processing**: Utilizes `OpenAI Whisper (Base)` to transcribe live microphone input into high-quality text, handling diverse accents and background noise. 2. **Intent & Entity Extraction**: Employs `Cohere's Command-R` to parse the user's speech and extract key entities like **Location**, **Hotel Category**, and **Preferences**. 3. **Reasone …

Visit

github.com

Tasks

automatic speech recognitionspeech processing