Spectral decomposition, as a special attribute, gives geologists and geophysicists a unique perspective for seeing and interpreting seismic data. The short-time Fourier transform (STFT) is similar to the internal workings of well-established, time-varying, spectral-balancing algorithms used in seismic processing. The continuous wavelet transform (CWT) is an alternative approach and has proven to be a better choice than the STFT for broadband data. As an extension of the CWT, I introduce the continuous wavelet packet-like transform (CWPT). The CWPT is flexible, and has high resolution if it is implemented properly. For interpretation using spectral decomposition with this method, the frequency absorption property is more accurately represented, and thus is a potential technique for gas reservoir interpretation.