Various decomposition techniques have been employed in signal processing for exploiting and high-lighting the characteristics of a given signal. In this paper we present orthogonal overlapping block transforms as a signal analysis tool with the capability of variable multiresolution time-spectral decomposition of speech signals. Our prime interest is in the representation of nonstationary discrete-time signals in terms of wavelet packets used as time-varying discrete orthogonal systems, and we concentrate on the fast transform algorithms for such systems. Much of our current knowledge and intuition of speech is derived from analysis involving assumptions of short-time stationarity (e.g., the DFT-based speech spectrogram). Such methods are, by their very nature, incapable of revealing the true nonstationary nature of speech. A careful consideration of the theory of time-varying orthogonal transforms, however, allows the construction of dynamic multiresolution spectrograms that reveal far more of the nonstationarities of speech, thereby highlighting just what it is that conventional approaches miss. The fast orthogonal overlapping block transform algorithms represent an elegant and efficient solution to the implementation of time-varying wavelet packet transforms, since their FFT-like lattice block structure provides all possible multiresolution time-spectral coefficients.