Accurately estimating state of charge (SOC), state of health (SOH), and other metrics of battery safety remains a challenge for lithium ion batteries outside the laboratory, where one may use techniques such as X-ray diffraction, electrochemical impedance spectroscopy, or neutron imaging. Existing in situ methods to estimate SOC and SOH employ voltage monitoring or Coulomb-counting. Unfortunately, these methods remain inaccurate because they assume a battery capacity or voltage per discharge, while capacity and voltage are known to vary with C-rate and cycle age. The present work explores the viability of interpreting SOC, SOH, and safety metrics by monitoring and interpreting ultrasonic signals propagating through a battery as it is cycled between charged and discharged states in various environmental conditions. Of specific interest are cases where a battery enters dangerous operating conditions which lead to the formation of gases within the battery as well as increases in internal temperature, pressure, and battery volume. To investigate this regime, the received ultrasonic signals are monitored over the course of charging and discharging cycles and analyzed in the time and frequency domains in an effort to provide an in situ means to estimate SOC, SOH and battery safety metrics.