Abdul-adeiza/naija-language-id

Naija-Language-ID: End-to-end AI system for detecting Nigeria’s major languages (Hausa, Igbo, Yoruba) from short audio clips — covering data collection, preprocessing, deep learning with PyTorch, and deployment. # 🎙️ Naija-Language-ID ## 📌 Project Title Naija-Language-ID: AI Model for Nigerian Local Language Detection 📖 Read the full story behind this project on Medium --- ## 📖 Description This project is an **end-to-end machine learning pipeline** designed to detect Nigeria’s three major local languages (**Hausa, Igbo, Yoruba**) from short audio clips. The workflow covers: - Data collection (YouTube) - Audio preprocessing (resampling, slicing, speech detection with Silero VAD) - Feature extraction and deep learning model training (PyTorch) - Deployment of a demo application for real-time inference The goal is to build a robust **speech classification system** that supports Nigeria’s diverse linguistic landscape and has applications in **customer service, fintech, telecom, and language technology**. --- ## ⚙️ Requirements Install dependencies using the provided `requirements.txt`: ```bash pip install -r requirements.txt