sharpness_din_from_loudness

sharpness_din_from_loudness(N, N_specific, weighting='din')

Compute the sharpness value from loudness

This function computes the sharpness value according to different methods.

Acoustic sharpness calculation according to different methods

(Aures, Von Bismarck, DIN 45692, Fastl) from loudness.

Parameters:

• N (float or array_like) – Overall loudness [sones], size (Ntime,).

• N_specific (numpy.ndarray) – Specific loudness array [sones/bark], size (Nbark, Ntime).

• weighting ({'din', 'aures', 'bismarck', 'fastl'}) – Weighting function used for the sharpness computation. Default is ‘din’

Returns:

• S (float or numpy.ndarray) – Sharpness value in [acum], dim (nseg)

• time_axis (numpy.array) – Time axis cut according to skip argument, size (Ntime,).

Warning

The sampling frequency of the signal must be >= 48 kHz to fulfill requirements. If the provided signal doesn’t meet the requirements, it will be resampled.

See also

sharpness_din_tv

Sharpness computation for a non-stationary time signal

sharpness_din_st

Sharpness computation for a stationary time signal

sharpness_din_freq

Sharpness computation from a sound spectrum

sharpness_din_perseg

Sharpness computation by time-segment

Notes

The computation consists of a specific loudness weighting employing a weighting function \(g(z)\):

\[S=0.11\frac{\int_{0}^{24Bark}N'(z)g(z)\textup{dz}}{N}\]

with \(N'\) the specific loudness and \(N\) the global loudness.

The different methods available with the function account for the weighting function applied:

• DIN 45692 : weighting defined in the standard

• Aures

• Bismarck

• Fastl

References

[1]

DIN.45692:2009. Measurement technique for the simulation of the auditory sensation of sharpness. Deutsches Institut fur Normung, 2009. URL: https://www.normadoc.com/english/din-45692-2009-08.html?___from_store=french.

[2]

E.Zwicker and H.Fastl. Psychoacoustics. Springer Berlin, 2007. URL: https://link.springer.com/book/10.1007/978-3-540-68888-4.

[3]

E. von Bismarck. Sharpness as an attribute of the timbre of steady sounds. Acta Acustica, 1974. URL: https://www.ingentaconnect.com/contentone/dav/aaua/1974/00000030/00000003/art00006.

Examples

>>> from mosqito.sq_metrics import loudness_zwtv, sharpness_din_from_loudness
>>> import matplotlib.pyplot as plt
>>> import numpy as np
>>> f=1000
>>> fs=48000
>>> d=0.2
>>> dB=60
>>> time = np.arange(0, d, 1/fs)
>>> f = np.linspace(1000,5000, len(time))
>>> stimulus = 0.5 * (1 + np.sin(2 * np.pi * f * time))
>>> rms = np.sqrt(np.mean(np.power(stimulus, 2)))
>>> ampl = 0.00002 * np.power(10, dB / 20) / rms
>>> stimulus = stimulus * ampl
>>> N, N_spec, bark_axis, time_axis = loudness_zwtv(stimulus, fs=fs)
>>> S = sharpness_din_from_loudness(N, N_spec)
>>> plt.plot(time_axis, S)
>>> plt.xlabel("Time [s]")
>>> plt.ylabel("Sharpness [Acum]")

(Source code, png, hires.png, pdf)

Warning

Note the transient effect at the beginning of the signal.