This interactive demo showcases how acoustic and vibration signal processing combined with machine learning can objectively characterize food texture. Developed within the TETRA KRAK project by VIVES University of Applied Sciences and KU Leuven, the platform captures sound, vibration and force signals as food products such as chips, croissants, biscuits, breaded products and Belgian chocolate are broken under controlled conditions. Real-time signal analysis and machine-learning models link these physical measurements to sensory attributes such as crispness, crunchiness or snap, complementing traditional sensory panel assessments. Visitors can observe live experiments, listen to fracture sounds, explore signal visualizations and compare texture differences, discovering new possibilities for objective food quality assessment and product development.