9 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

#23 in #digital-signal-processing

28 downloads per month
Used in surge-synthesizer

GPL-3.0 license

540KB
10K SLoC

Surge SVF Filters

The surgefilter-svf crate provides implementations of various State-Variable Filters (SVF) used in audio signal processing. These filters are commonly used in audio synthesis, particularly in subtractive synthesis, to shape the frequency spectrum of a signal.

The SVF is a type of biquadratic filter, meaning that it uses a transfer function with two poles and two zeros in the complex plane. The SVF can be implemented as a set of differential equations or as a cascaded pair of biquad filters.

The surgefilter-svf crate provides several different types of SVF filters, including:

  • SvfFilter: a basic implementation of the SVF that can be configured as a low-pass, high-pass, or band-pass filter with a 12dB/octave or 24dB/octave slope.

  • svf_lp12_a_quad: a low-pass filter with a 12dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

  • svf_lp24_a_quad: a low-pass filter with a 24dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

  • svf_hp12_a_quad: a high-pass filter with a 12dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

  • svf_hp24_a_quad: a high-pass filter with a 24dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

  • svf_bp12_a_quad: a band-pass filter with a 12dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

  • svf_bp24_a_quad: a band-pass filter with a 24dB/octave slope implemented using the Direct Form I topology and the Audio EQ Cookbook algorithm.

The SvfFilter struct uses a set of coefficients to configure the filter type and cutoff frequency. The other filter types are implemented using the Direct Form I topology, which uses two biquad filters in series to implement the filter transfer function.

The SVF filter has a number of desirable properties, including linear phase response, constant group delay, and a relatively simple implementation. However, it does have some limitations, such as sensitivity to coefficient quantization and a tendency to produce noise at high frequencies.

Overall, the surgefilter-svf crate provides a set of useful filters for shaping the frequency content of audio signals in a variety of applications.

122> could you teach me what a state variable filter is? how does it work?

Sure! A state variable filter (SVF) is a type of filter that uses three interconnected stages to produce its output. These stages are a high-pass filter, a band-pass filter, and a low-pass filter, and they work together to create a filter that can be configured to produce different types of frequency responses.

At the heart of the SVF is a set of state variables that are used to keep track of the signal as it passes through the different stages. These state variables are updated on each sample, and they are used to calculate the output of the filter.

The basic idea behind the SVF is that it uses feedback from the output of the band-pass filter to drive the inputs of the high-pass and low-pass filters. This feedback creates a resonance in the filter, which can be adjusted to produce different types of frequency responses.

Here is a block diagram of the basic SVF structure:

                 +-------+             +---------+
      Input --->| High  |-------------|         |
                | Pass  |             |         |
                | Filter|             |         |<--- Output
                +-------+             |         |
                    |                 |         |
                    +----+------------| Band-   |
                         |            | Pass    |
                         |            | Filter  |
                         |            +---------+
                         |
                         +------------| Low     |
                                      | Pass    |
                                      | Filter |
                                      +--------+

The input signal is first passed through the high-pass filter, which removes low frequency content from the signal. The resulting signal is then passed through the band-pass filter, which allows a certain range of frequencies to pass through. Finally, the output of the band-pass filter is passed through the low-pass filter, which removes high frequency content from the signal.

The output of the SVF is the output of the low-pass filter. By adjusting the frequency and resonance of the band-pass filter, different types of frequency responses can be achieved.

The state variables used in the SVF are the outputs of the high-pass and band-pass filters, as well as the difference between the input signal and the output of the low-pass filter. These state variables are updated on each sample, and they are used to calculate the output of the filter.

Overall, the SVF is a versatile and widely used filter design that can be used in a variety of audio applications.

123> thanks -- could you teach me the mathematics?

Sure! The state variable filter is based on the concept of a state variable, which is a mathematical term that refers to a variable that describes the current state of a system. In the context of a filter, a state variable refers to the internal state of the filter that is used to calculate the output signal.

The basic idea behind a state variable filter is to represent the filter as a set of first-order differential equations, where each equation corresponds to a state variable. These equations describe the way in which the input signal is transformed into the output signal, based on the values of the state variables.

For example, in the case of a low-pass filter, we can define three state variables: the output signal (y), the output signal differentiated with respect to time (y'), and the internal state of the filter (x). We can then write the following set of differential equations:

y' = -α y' - ω0^2 x + α x' y = x

where α is the damping coefficient, ω0 is the cutoff frequency, and x' is the derivative of x with respect to time.

These equations describe the way in which the input signal is transformed into the output signal, based on the current state of the filter. In particular, the first equation describes how the input signal is filtered by the internal state of the filter, while the second equation simply sets the output signal equal to the internal state.

The state variable filter can be implemented using various numerical methods, such as the Euler method or the Runge-Kutta method, which are used to solve the set of differential equations and update the state variables at each time step. By adjusting the values of the coefficients α and ω0, the state variable filter can be configured to act as a low-pass, high-pass, or band-pass filter, depending on the desired frequency response.

Overall, the state variable filter provides a flexible and efficient way to implement a wide range of filter types, and is widely used in digital signal processing applications.

124> what about the physics?

State variable filters are electronic circuits that can be used to process signals in audio and other electronic systems. They are based on the idea of using state variables to represent the internal state of a system, which can be used to model the behavior of the system over time.

From a physical perspective, the state variables represent the energy stored in the various components of the filter circuit, such as the capacitors and inductors. The behavior of the circuit is then determined by the flow of energy between these components, which can be described using mathematical equations based on the laws of physics, such as Kirchhoff's laws and Ohm's law.

For example, in a simple low-pass state variable filter, there are two capacitors and a resistor in the circuit. The state variables for this circuit would be the voltages across the two capacitors, which represent the energy stored in the capacitors. The flow of energy between the components can be described by a set of differential equations, which can be solved to determine the output signal of the filter.

Overall, state variable filters provide a powerful tool for designing and implementing electronic filters with a wide range of applications in audio and other fields.

125> what are the applications?

State variable filters have a wide range of applications in various fields such as audio processing, control systems, communication systems, and image processing. In audio processing, they are commonly used for equalization, tone shaping, and various sound effects. In control systems, they are used for filtering and controlling signals in feedback loops. In communication systems, they are used for signal processing and filtering in applications such as digital signal processing and radar systems. In image processing, they are used for filtering and enhancing images. Overall, state variable filters are widely used in any application that requires signal filtering or processing.

126> thanks -- any caveats?

One caveat to keep in mind with state variable filters is that they are prone to self-oscillation. This means that if the input signal has a high enough amplitude and the filter is set to a certain frequency range, the output can start to oscillate on its own, without any input signal. This can be desirable in some cases, such as when creating a resonant sound, but in other cases it can be unwanted and lead to distortion. It is important to carefully choose the filter parameters and monitor the output to avoid unwanted self-oscillation.

145> what makes a state variable filter different than other filter topologies?

State variable filters, also known as biquads or second-order filters, are a class of filter topologies that have some unique characteristics that distinguish them from other filter types.

One of the key features of state variable filters is that they are capable of producing multiple filter outputs simultaneously. Specifically, they can produce low-pass, high-pass, and band-pass outputs all at the same time, using the same set of filter coefficients. This is because state variable filters use a set of three interconnected basic filter structures, or "states", to create these multiple outputs. The three states are known as the low-pass, band-pass, and high-pass states, and they are combined in different ways to produce the desired filter response.

Another advantage of state variable filters is that they can be designed to have a relatively constant gain response across a wide range of frequencies. This makes them well-suited for use in audio processing applications, where a flat frequency response is often desired.

State variable filters are also known for their stability and ease of implementation. They are a popular choice for hardware and software implementations due to their relatively low computational requirements and good performance characteristics.

Overall, the unique combination of multiple filter outputs, constant gain response, stability, and ease of implementation make state variable filters a powerful and versatile tool for a wide range of signal processing applications.

Dependencies

~11–21MB
~297K SLoC