Our solution tackles the inefficiencies in battery formation by introducing real- time monitoring and quality control, significantly reducing waste and improving production speed. We achieve this through an acoustic sensor placed on the battery during the formation process. This sensor detects the sounds of gasing and cracking inside the cell in real time, allowing us to track the electrochemical changes occurring inside the cell. Instead of waiting until the end of the process to detect defects, our system analyzes these signals immediately using a machine learning (ML) model. With these insights, our optimized protocol reduces the manufacturing time by over 1.5 weeks (~40% reduction) while enabling real-time predictive quality assurance (QA/QC). This prevents defective batteries from progressing through production, allowing early adjustments to the production protocol and reducing scrap rate significantly.