In a world where voice-based technologies are rapidly evolving, the rise of AI-generated synthetic voices poses serious concerns for authenticity, privacy, and security. Echo Scan is a machine learning-based system designed to differentiate between real human voices and artificially generated ones. The system leverages acoustic features, waveform analysis, and deep learning techniques to identify subtle inconsistencies that are often overlooked by the human ear. Through extensive training on diverse datasets and voice patterns, Echo Scan aims to act as a reliable shield against voice cloning, fraud, and misinformation. This project not only strengthens digital trust but also sets the foundation for future advancements in audio forensics and secure voice authentication.