real_time_fir_iir_filters

Ever needed a low pass filter for your VST? This crate has a wide selection of filters for real-time usage. It's designed to have as little runtime overhead as possible.

How does it work?

Everything that can be computed at compile-time, will be, and the filter coefficients will be cached as well.

I then use the following algorithm to process the signal with as few steps as possible given the filter's coefficients:

(The figure is from: Alan V. Oppenheimer & Ronald W. Schafer - Discrete-Time Signal Processing)

Example

use core::f64::consts::TAU;

use real_time_fir_iir_filters::{
    conf::LowPass,
    param::OmegaEpsilonXi,
    rtf::Rtf,
    filters::iir::second::SecondOrderEllipticFilter
};

// Initialize a 2. order elliptic low-pass filter at 440Hz
let mut filter = SecondOrderEllipticFilter::new::<LowPass>(
    OmegaEpsilonXi {
        omega: 440.0*TAU,
        epsilon: 0.5,
        xi: 1.5
    }
);

const N: usize = 10;
const RATE: f64 = 8000.0;

{
    // Unit impulse
    let mut imp_resp = [0.0; N];
    imp_resp[0] = 1.0;

    // Apply filter to imp_resp
    for x in &mut imp_resp
    {
        [*x] = filter.filter(RATE, *x);
    }

    // Prints the impulse response of the filter.
    println!("h[n] = {:?}", imp_resp);
}

// Resets internal state of filter to zero
filter.reset();

{
    // Generate frequency response for ω ∈ [0, 2π)
    let freq_resp = core::array::from_fn::<_, N, _>(|i| {
        let omega = i as f64/N as f64*TAU;

        filter.frequency_response(RATE, omega)
    });

    println!("H = {:?}", freq_resp);
}

Available filters

...and more to come!

Adding your own filter

You can also implement your own filter, by using the macro def_rtf!. See how i did it with the other filters for an example on how to use the macro.