An algorithm for estimating signals from short-time magnitude spectra is introduced offering a significant improvement in quality and efficiency over current methods. The key issue is how to invert a sequence of overlapping magnitude spectra (a ldquospectrogramrdquo) containing no phase information to generate a real-valued signal free of audible artifacts. Also important is that the algorithm performs in real-time, both structurally and computationally. In the context of spectrogram inversion, structurally real-time means that the audio signal at any given point in time only depends on transform frames at local or prior points in time. Computationally, real-time means that the algorithm is efficient enough to run in less time than the reconstructed audio takes to play on the available hardware. The spectrogram inversion algorithm is parameterized to allow tradeoffs between computational demands and the quality of the signal reconstruction. The algorithm is applied to audio time-scale and pitch modification and compared to classical algorithms for these tasks on a variety of signal types including both monophonic and polyphonic audio signals such as speech and music.

In this paper, we present an algorithm for real-time iterative spectrogram inversion (RTISI) with look-ahead (RTISI-LA). RTISI-LA reconstructs a time-domain signal from a given sequence of short-time Fourier transform magnitude (STFTM) spectra without phase information. Whereas RTISI reconstructs the current frame using only magnitude spectra information for previous frames and the current frame, RTISI-LA also uses magnitude spectra for a small number future frames. This allows RTISI-LA to achieve substantially higher signal-to-noise (SNR) performance than either RTISI or the Griffin & Lim method with an equivalent computational load, while retaining the real-time properties of RTISI