Electric vehicles (EVs) are essential for future transportation to help in reducing environmental emissions. The main and most critical component of these vehicles is the traction motors, which power the wheels and propel the vehicle forward. However, these traction motors are susceptible to faults like any other machine that can impact the performance and safety of the vehicle. This research proposes a Wavelet-Based Neural Network (WBNN) diagnostic methodology that can diagnose the faults of PMSM and BLDC motors. Initially, several experiments were conducted on both motors based on different operating and speed conditions. Data has been collected through a vibration and current sensor in each operating condition during this experimentation. Thereafter, the raw signal has been pre-processed through wavelet analysis and converted into time-frequency spectrums to further enhance the fault characteristic. Finally, an optimized neural network has been employed to further categorize the faults in both motors. The validation of the proposed diagnostic methodology has been validated on the actual electric vehicles on both the PMSM and BLDC motors. This developed diagnostic strategy can help in the timely detection of faults and abnormalities in electric vehicle motors to ensure reliable performance, prevent safety hazards, and minimize maintenance costs.