In this study, a new analysis method using a wavelet transform was considered to evaluate the chatter vibration generated during end milling. End milling often generates vibrations between the tool and work material, called chatter vibration, which causes deterioration of the finished surface and breakage of the tool. Therefore, countermeasures to detect chatter vibration at an early stage have been attempted in the past by using fast Fourier transform (FFT) and short-time Fourier transform (STFT) methods and monitoring the dynamic stability of the cutting process. However, the FFT analysis method assumes steady-state vibration, and the STFT method does not have sufficient frequency resolution. In contrast, the wavelet transform is excellent for analyzing non-stationary vibrations and has a high noise separation capability. To fully validate the analysis method, a groove was added to the machined surface, so that the cutting condition changed with time, and the cutting vibration under the condition where the disturbance was involuntary was analyzed. As a result, it was possible to identify minute fluctuations in chatter vibration, which could not be obtained using the STFT method.