8 releases
0.2.12-alpha.0 | Apr 7, 2023 |
---|---|
0.2.11-alpha.0 | Dec 19, 2022 |
0.2.5-alpha.0 | Jun 21, 2022 |
0.2.4-alpha.0 | Mar 14, 2022 |
0.1.42-alpha.0 | Oct 27, 2021 |
#5 in #tangent
40 downloads per month
Used in surge-synthesizer
155KB
2.5K
SLoC
TanhShaper
The TanhShaper
is an implementation of a soft
clipping algorithm using the hyperbolic tangent
function. This shaper can be used to add
saturation and warmth to an audio signal, similar
to a tube amplifier. The amount of shaping can be
controlled by adjusting the shape
parameter.
Mathematical analysis
The hyperbolic tangent function is defined as:
tanh(x) = (exp(x) - exp(-x)) / (exp(x) + exp(-x))
When x
is small, tanh(x)
is approximately
equal to x
, and as x
becomes larger, tanh(x)
approaches 1. This non-linear response curve makes
it suitable for use as a shaper function.
The TanhShaper
takes an input signal and applies
the hyperbolic tangent function to it, which has
the effect of softly clipping the signal. The
degree of soft clipping is controlled by the
shape
parameter, which scales the input signal.
Usage
The TanhShaper
can be used in any application
where saturation or warmth is desired, such as in
guitar amplifiers, synthesizers, or mixing
consoles. It can be used in conjunction with other
shaping algorithms to create complex and unique
sounds.
Caveats
One potential issue with using a soft clipping
algorithm like TanhShaper
is that it can
introduce additional harmonic distortion to the
signal. This can result in a loss of clarity or
fidelity, depending on the
application. Additionally, care must be taken when
adjusting the shape
parameter to avoid
introducing unwanted artifacts or distortion.
Overall, the TanhShaper
is a powerful tool for
shaping audio signals and can be a valuable
addition to any audio processing toolbox.
94> could you teach me about the hyperbolic tangent function?
Sure! The hyperbolic tangent function, usually
denoted as tanh(x)
, is a mathematical function
that maps any real number x
to a value between
-1 and 1. The function is defined as:
tanh(x) = (e^x - e^-x) / (e^x + e^-x)
where e
is the mathematical constant
approximately equal to 2.71828.
The graph of the hyperbolic tangent function is
similar to the graph of the standard tangent
function, but with a vertical scaling of 1 and
a horizontal scaling of 2. The function is
symmetric about the origin, meaning that
tanh(-x) = -tanh(x)
for all x
.
The hyperbolic tangent function is often used as an activation function in artificial neural networks because it is differentiable and has a range between -1 and 1. It is also commonly used in signal processing, particularly in audio and image processing, for its ability to compress and expand dynamic range.
95> could you show me some python code illustrating the effect on the input signal?
Certainly, here is an example of using the tanh
function to shape an input signal:
import numpy as np
import matplotlib.pyplot as plt
# Create a sine wave input signal
sample_rate = 44100
duration = 1.0
freq = 440.0
t = np.linspace(0, duration, int(sample_rate * duration), endpoint=False)
signal = np.sin(2 * np.pi * freq * t)
# Apply tanh shaping with a shape factor of 0.5
shape_factor = 0.5
shaped_signal = np.tanh(signal * shape_factor)
# Plot the input and output signals
fig, ax = plt.subplots(2, 1)
ax[0].plot(t, signal)
ax[0].set_title('Input Signal')
ax[0].set_xlabel('Time (s)')
ax[0].set_ylabel('Amplitude')
ax[1].plot(t, shaped_signal)
ax[1].set_title('Output Signal')
ax[1].set_xlabel('Time (s)')
ax[1].set_ylabel('Amplitude')
plt.show()
In this example, we create a 1 second sine wave
input signal with a frequency of 440 Hz sampled at
a rate of 44100 Hz. We then apply the tanh
function with a shape factor of 0.5 to the input
signal to create the shaped output signal. The
resulting input and output signals are then
plotted for visualization.
Dependencies
~12–20MB
~302K SLoC