VocalID is an open-source Python framework for speaker verification that provides a lightweight, transparent, and accessible approach to voice-based biometric authentication. The system combines a pretrained ECAPA-TDNN speaker embedding model with a logistic regression decision layer to deliver accurate speaker verification without requiring GPU hardware or complex deployment pipelines. Designed for researchers, educators, students, and practitioners, VocalID supports both file-based and real-time microphone verification through a Python API, command-line interface, and optional FastAPI server. The framework emphasizes modularity and reproducibility, allowing users to easily experiment with different embedding models and classification strategies while maintaining a simple installation process. Experimental evaluation demonstrates competitive verification performance, achieving a 3.8% Equal Error Rate (EER) and an AUC of 0.974 on a controlled multi-speaker dataset. By separating embedding extraction from classification, VocalID provides an interpretable and extensible architecture suitable for educational use, rapid prototyping, and applied research in voice biometrics. The project is released under the MIT License and is available as a pip-installable package, lowering the barrier to entry for speaker verification research and practical deployment in resource-constrained environments.