invertSpectrogram: Invert spectrogram
In soundgen: Sound Synthesis and Acoustic Analysis

invertSpectrogram

R Documentation

Invert spectrogram

Description

Transforms a spectrogram into a time series with inverse STFT. The problem is that an ordinary spectrogram preserves only the magnitude (modulus) of the complex STFT, while the phase is lost, and without phase it is impossible to reconstruct the original audio accurately. So there are a number of algorithms for "guessing" the phase that would produce an audio whose magnitude spectrogram is very similar to the target spectrogram. Useful for certain filtering operations that modify the magnitude spectrogram followed by inverse STFT, such as filtering in the spectrotemporal modulation domain.

Usage

invertSpectrogram(
  spec,
  samplingRate,
  windowLength,
  overlap,
  step = NULL,
  wn = "hanning",
  specType = c("abs", "log", "dB")[1],
  initialPhase = c("zero", "random", "spsi")[3],
  nIter = 50,
  normalize = TRUE,
  play = TRUE,
  verbose = FALSE,
  plotError = TRUE
)

Arguments

`spec`	the spectrogram that is to be transform to a time series: numeric matrix with frequency bins in rows and time frames in columns
`samplingRate`	sampling rate of `x` (only needed if `x` is a numeric vector)
`windowLength`	length of FFT window, ms
`overlap`	overlap between successive FFT frames, %
`step`	you can override `overlap` by specifying FFT step, ms (NB: because digital audio is sampled at discrete time intervals of 1/samplingRate, the actual step and thus the time stamps of STFT frames may be slightly different, eg 24.98866 instead of 25.0 ms)
`wn`	window type accepted by `ftwindow`, currently gaussian, hanning, hamming, bartlett, blackman, flattop, rectangle
`specType`	the scale of target spectroram: 'abs' = absolute, 'log' = log-transformed, 'dB' = in decibels
`initialPhase`	initial phase estimate: "zero" = set all phases to zero; "random" = Gaussian noise; "spsi" (default) = single-pass spectrogram inversion (Beauregard et al., 2015)
`nIter`	the number of iterations of the GL algorithm (Griffin & Lim, 1984), 0 = don't run
`normalize`	if TRUE, normalizes the output to range from -1 to +1
`play`	if TRUE, plays back the reconstructed audio
`verbose`	if TRUE, prints estimated time left every 10% of GL iterations
`plotError`	if TRUE, produces a scree plot of squared error over GL iterations (useful for choosing 'nIter')

Details

Algorithm: takes the spectrogram, makes an initial guess at the phase (zero, noise, or a more intelligent estimate by the SPSI algorithm), fine-tunes over 'nIter' iterations with the GL algorithm, reconstructs the complex spectrogram using the best phase estimate, and performs inverse STFT. The single-pass spectrogram inversion (SPSI) algorithm is implemented as described in Beauregard et al. (2015) following the python code at https://github.com/lonce/SPSI_Python. The Griffin-Lim (GL) algorithm is based on Griffin & Lim (1984).

Value

Returns the reconstructed audio as a numeric vector.

References

Griffin, D., & Lim, J. (1984). Signal estimation from modified short-time Fourier transform. IEEE Transactions on Acoustics, Speech, and Signal Processing, 32(2), 236-243.
Beauregard, G. T., Harish, M., & Wyse, L. (2015, July). Single pass spectrogram inversion. In 2015 IEEE International Conference on Digital Signal Processing (DSP) (pp. 427-431). IEEE.

Examples

# Create a spectrogram
samplingRate = 16000
windowLength = 40
overlap = 75
wn = 'gaussian'

s = soundgen(samplingRate = samplingRate, addSilence = 100)
spec = spectrogram(s, samplingRate = samplingRate,
  wn = wn, windowLength = windowLength, step = NULL, overlap = overlap,
  padWithSilence = FALSE, output = 'original')

# Invert the spectrogram, attempting to guess the phase
# Note that samplingRate, wn, windowLength, and overlap must be the same as
# in the original (ie you have to know how the spectrogram was created)
s_new = invertSpectrogram(spec, samplingRate = samplingRate,
  windowLength = windowLength, overlap = overlap, wn = wn,
  initialPhase = 'spsi', nIter = 100, specType = 'abs', play = FALSE)

# Verify the quality of audio reconstruction
# playme(s, samplingRate); playme(s_new, samplingRate)

soundgen documentation built on Sept. 12, 2024, 6:29 a.m.