A Fast DSP library for effect and synthesizer development in C. Optimized for iOS/OSX using the Accelerate framework but should run using non-vectorized implementations for use on any platform.

FxDSP can be built and installed on most platforms using CMake

mkdir build
cd build
cmake ../FxDSP
make && make install

On Mac OS X, FxDSP can be built and installed using the included Xcode Project

Create a low-shelf filter and use it to process some samples:

#include "FxDSP/RBJFilter.h"    

unsigned num_samples;

// Single Precision example ------------------------------------------
float* input_samples;
float* output_samples;

// Initialize the Filter
RBJFilter *filter = RBJFilterInit(LOW_SHELF, 44100, 100);

// Process some data
RBJFilterProcess(filter, output_samples, input_samples, num_samples);

// Cleanup
RBJFilterFree(filter);

// Double Precision example ------------------------------------------
double* input_samplesD;
double* output_samplesD;

// Initialize the Filter
RBJFilterD *filter = RBJFilterInitD(LOW_SHELF, 44100, 100);

// Process some data
RBJFilterProcessD(filter, output_samplesD, input_samplesD, num_samples);

// Cleanup
RBJFilterFreeD(filter);

A Circular buffer with fast wrapping.

Digital Biquad filter. A building block for more complex filters. Uses a vectorized implementation if the Accelerate framework is available, otherwise uses a stil-pretty-efficient DF-II implementation

FIR Filter for processing audio with arbitrary FIR filters. Under the hood, the filter uses vectorized convolution for short kernels, and FFT convolution for fast processing with larger kernels.

Biquad EQ stages, implements shelves, low/high-pass, bandpass, and notch filters with adjustable cutoff and Q. Each stage is a single biquad section. Chain a few of these stages together to create a parametric EQ.

A single pole filter, good for anywhere you need a quick and easy way to roll off some lows/highs, such as for band-limiting control parameter changes. They can also be used to generate envelope attack/decay/release curves if you use it to filter an envelope defined as a bunch of step changes.

A Multiband filter bank. Splits the signal into 3 bands (low, mid, and high), that mix back flat. The split points can be set anywhere.

Fast Fourier Transforms. Allows for processing audio in the frequency domain. The FFT functions abstract away most of the setup, real/complex packing, etc. that you've probably come to expect from an FFT library so you can focus on making cool plugins/visualizers/etc. without having to go digging through your textbooks to refresh your memory on twiddle factors.

Window functions. Blackman, Kaiser, Hamming, Hann and many more. Used for removing edge effects from a sliding window when doing frequency-domain analysis, eg. for a spectrum analyzer.

Digital implementation of a MOOG Ladder filter with adjustable cutoff and resonance.