SIGNAL RECONSTRUCTION FROM STFT MAGNITUDE : A STATE OF THE ART

This paper presents a review on techniques for signal reconstruction without phase, i.e. when only the spectrogram (the squared magnitude of the Short Time Fourier Transform) of the signal is known. The now standard Griffin and Lim algorithm will be presented, and compared to more recent blind techniques. Two important issues are raised and discussed: first, the definition of relevant criteria to evaluate the performances of different algorithms, and second the question of the unicity of the solution. Some ways of reducing the complexity of the problem are presented with the injection of additional information in the reconstruction. Finally, issues that prevents optimal reconstruction are examined, leading to a discussion on what seem the most promising approaches for future research.

[1]  R. Gerchberg.  A practical algorithm for the determination of phase from image and diffraction plane pictures , 1972 .

[2]  M. Portnoff,et al.  Implementation of the digital phase vocoder using the fast Fourier transform , 1976 .

[3]  F. Harris.  On the use of windows for harmonic analysis with the discrete Fourier transform , 1978, Proceedings of the IEEE.

[4]  Michael R. Portnoff.  Magnitude-phase relationships for short-time Fourier transforms based on Gaussian analysis windows , 1979, ICASSP.

[5]  A. Oppenheim,et al.  Signal reconstruction from phase or magnitude , 1980 .

[6]  Thomas F. Quatieri,et al.  The importance of boundary conditions in the phase retrieval problem , 1982, ICASSP.

[7]  Jae Lim,et al.  Signal reconstruction from short-time Fourier transform magnitude , 1983 .

[8]  A. Oppenheim,et al.  Signal reconstruction from signed Fourier transform magnitude , 1983 .

[9]  B. Yegnanarayana,et al.  Significance of group delay functions in signal reconstruction from spectral magnitude or phase , 1984 .

[10]  James R. Fienup,et al.  Phase-retrieval stagnation problems and solutions , 1986 .

[11]  Monson H. Hayes,et al.  Phase retrieval using a window function , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  Brendan J. Frey,et al.  Probabilistic Inference of Speech Signals from Phaseless Spectrograms , 2003, NIPS.

[13]  Lonce Wyse,et al.  AN EFFICIENT ALGORITHM FOR REAL-TIME SPECTROGRAM INVERSION , 2005 .

[14]  Kuldip K. Paliwal,et al.  On the usefulness of STFT phase spectrum in human listening tests , 2005, Speech Commun..

[15]  Tony Ezzat,et al.  Max-Gabor analysis and synthesis of spectrograms , 2006, INTERSPEECH.

[16]  Lonce L. Wyse,et al.  Real-Time Iterative Spectrum Inversion with Look-Ahead , 2006, 2006 IEEE International Conference on Multimedia and Expo.

[17]  Rémi Gribonval,et al.  Audio source separation with a single sensor , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[18]  Tony Ezzat,et al.  An incremental algorithm for signal reconstruction from short-time fourier transform magnitude , 2006, INTERSPEECH.

[19]  Kuldip K. Paliwal,et al.  Iterative reconstruction of speech from short-time Fourier transform phase and magnitude spectra , 2007, Comput. Speech Lang..

[20]  Lonce L. Wyse,et al.  Real-Time Signal Estimation From Modified Short-Time Fourier Transform Magnitude Spectra , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[21]  Jonathan Le Roux,et al.  Explicit consistency constraints for STFT spectrograms and their application to phase reconstruction , 2008, SAPA@INTERSPEECH.

[22]  Bin Yang.  A study of inverse short-time fourier transform , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[23]  Stéphane Mallat,et al.  Audio Denoising by Time-Frequency Block Thresholding , 2008, IEEE Transactions on Signal Processing.

[24]  Volker Gnann,et al.  Inversion of short-time fourier transform magnitude spectrograms with adaptive window lengths , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[25]  Hirokazu Kameoka,et al.  Consistent Wiener Filtering: Generalized Time-Frequency Masking Respecting Spectrogram Consistency , 2010, LVA/ICA.

[26]  Radu V. Balan,et al.  On signal reconstruction from its spectrogram , 2010, 2010 44th Annual Conference on Information Sciences and Systems (CISS).

[27]  Hirokazu Kameoka,et al.  Phase initialization schemes for faster spectrogram-consistency-based signal reconstruction ∗ ◎ , 2010 .

[28]  Volker Gnann,et al.  IMPROVING RTISI PHASE ESTIMATION WITH ENERGY ORDER AND PHASE UNWRAPPING , 2010 .

[29]  Bertrand Nouvel.  A study of a local-features-aware model for the problem of phase reconstruction from the magnitude spectrogram , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[30]  Jonathan Le Roux,et al.  FAST SIGNAL RECONSTRUCTION FROM MAGNITUDE STFT SPECTROGRAM BASED ON SPECTROGRAM CONSISTENCY , 2010 .

[31]  Deep Sen,et al.  Iterative Phase Estimation for the Synthesis of Separated Sources From Single-Channel Mixtures , 2010, IEEE Signal Processing Letters.