EmoLock is a smart security system that uses voice and emotion recognition for personnal identification. It could be used by companies for their employees' authentication. User story: an employee arrives at his office's door and is asked to prononce a random script with a specific emotion (sad, happy, angry, ...). The voice is processed through an AI system that performs: - record & audio capture - audio filtering & speech detection - voice's fingerprint extraction through a neural network - voice and emotion recognition (comparison to users' database - clusters) We created a simple and intuitive GUI and the whole solution runs into a Docker container. This project was realized for OTO.ai, a swiss-based startup that leverages Deep Learning techniques to extract information from human's voice.