The dispersion and multiple modes which exist in the Lamb waves propagating in metal plates make it extremely hard to analyze and recognize the detection and echo signals. Time-frequency analysis is one of the powerful tools to deal with the non-stationary signal, which can reveal the time-varying spectral characteristics of signals. The experimental platform with defected single aluminum plate coupling in the vertical way is setup and the detected Lamb wave signals are analyzed by the short-time Fourier transform (STFT) method. The dispersion characteristics are analyzed. By extracting and analyzing the excited frequency corresponded component in the time-frequency domain, the aliasing Lamb wave testing signals in the time domain are recognized. The results show that STFT can achieve a good analysis for the Lamb wave testing signals of metal plate’s defects, and by extracting some particular frequency component in the time-frequency domain, the aliasing Lamb wave testing signals of defects in the time domain can be recognized.