Sound Synthesis, Sampling & Digital Signal Processing
An interactive journey from oscillators and sampling to filters, reverberation, nonlinear systems and spectral processing. Each page introduces a concept through explanation, visualisation and hands-on experimentation.
Sound Synthesis & Instrument Building
Begin with oscillators and the basic ingredients of synthesis, then combine envelopes, modulation, sequencing, polyphony and MIDI into playable instruments.
Synthesis Prologue
→Start with the central idea of synthesis: creating and controlling sound from fundamental building blocks.
Synthesis & Aliasing
→Generate basic waveforms and hear what changes when harmonics cross the Nyquist limit.
The Big Three
→Explore the three core ingredients of a synthesiser: oscillator, filter and amplifier.
FM Synthesis
→Use one oscillator to modulate another and explore how frequency modulation creates new spectra — and aliasing.
ADSR Envelopes
→Shape sound through time using attack, decay, sustain and release.
Filter Envelopes
→Apply envelopes to timbre as well as amplitude and hear a filter evolve through a note.
LFO Modulation
→Use slow oscillators as control signals for vibrato, tremolo and other repeating movement.
Arpeggiators
→Turn held notes into patterns and explore musical events distributed through time.
Polyphony & Voice Allocation
→See how a synthesiser manages multiple simultaneous notes by allocating independent voices.
MIDI Messages
→Connect note number, velocity and controller data to the parameters of an instrument.
Drum Synthesis
→Build percussion from synthesis primitives and discover how envelopes, noise and pitch create drum-like sounds.
Step Sequencing
→Place synthesised percussion on a rhythmic grid and turn the individual sounds into a simple drum machine.
Sampling & Alternative Synthesis
Replace the oscillator with recorded sound, then explore the many ways a buffer can become an instrument: transposition, interpolation, stretching, grains and wavetables.
Sampling — Audio in a Buffer
→Store recorded audio in memory and treat the buffer as a new source for synthesis.
Sampling — Key Spanning & Editing
→Trim, reverse and transpose samples across a keyboard while exploring the consequences of basic resampling.
Interpolation — Reading Between Samples
→Explore what happens when playback positions fall between stored samples, from simple interpolation toward sinc reconstruction.
Time Stretching & Pitch Shifting
→Separate two properties that simple resampling couples together: playback duration and pitch.
Granular Synthesis
→Break a buffer into short grains and use timing, position and stereo spread to create evolving textures.
Wavetable Synthesis
→Shrink the buffer concept until a tiny repeating waveform effectively becomes an oscillator.
BONUSPhysical Modelling
→Explore synthesis based on the behaviour of a resonating physical system rather than a conventional oscillator.
BONUSScrub & Pitch
→Treat playback position itself as a performance control and explore the relationship between scrubbing, speed and pitch.
Putting the Instrument Together
→Bring the synthesis, sampling and control concepts together into a more complete playable instrument.
DSP Fundamentals
Look beneath the instruments and effects. These experiments reduce DSP to reusable operations: multiply, store, delay, add, filter, feed back and combine.
The DSP Toolbox
→Map the small set of mathematical and memory operations that recur throughout digital audio processing.
Multiply — Gain & Modulation
→See how one simple operation controls level and becomes the basis of amplitude modulation and many other processes.
Store & Delay — DSP Memory
→Introduce sample memory and discover how storing previous samples creates delay and state.
Modulated Delay — Flanging & Chorus
→Move a delay through time and hear how the same structure becomes flanging, chorus and animated filtering.
Inside the EQ
→Open up an equaliser and connect familiar frequency controls to the coefficients doing the sample-by-sample calculation.
How Averaging Becomes Low-pass
→Average neighbouring samples and watch a simple arithmetic operation turn into a frequency-selective filter.
Average vs Difference
→Compare addition and subtraction of neighbouring samples to reveal simple low-pass and high-pass behaviour.
FIR Taps & Coefficients
→Build filters from delayed input samples, weights and summation, and experiment directly with FIR coefficients.
IIR — Feedback & Resonance
→Feed previous outputs back into a filter and discover how recursion introduces memory, resonance and potentially infinite decay.
Practical EQ — Plugin to DSP
→Connect the controls of everyday EQ practice to the filter structures and coefficients underneath them.
Allpass, Phase & Phaser
→Change phase without directly changing magnitude, then use that behaviour to understand phasing effects.
Reverb, Convolution & Acoustic Measurement
Follow delay and filtering into artificial space, then connect algorithmic reverberation to measured spaces through impulse responses and convolution.
From Echoes to Reverb
→Move from recognisable repetitions toward dense reflection patterns that are perceived as reverberation.
Algorithmic Reverb Laboratory
→Experiment with the structures and parameters that turn networks of delays and feedback into synthetic acoustic spaces.
Impulse Responses & Convolution
→Treat an acoustic response as a long FIR filter and use convolution to apply one system's behaviour to another signal.
Measuring an Impulse Response — ESS
→Explore exponential sine sweep measurement and how a real space or system can be captured as an impulse response.
BONUSCreative Impulse Responses
→Move beyond literal room capture and explore impulse responses as material for designed stereo motion and unusual spaces.
FIR vs IIR
→Compare feed-forward and feedback structures in terms of response, memory, efficiency and practical audio applications.
Dynamics, Nonlinearity & Modelling
Measure changing level, connect it to dynamics processing, then leave the linear world to explore distortion, aliasing, oversampling and different approaches to modelling real devices.
Peak, RMS & Envelope
→Compare ways of measuring signal level and see how an evolving amplitude envelope can be extracted from audio.
Dynamics — Connect the Wire
→Use level measurement to control gain and connect the building blocks behind compressors and other dynamics processors.
Distortion & Waveshaping
→Break linearity deliberately and see how changing waveform shape generates new harmonic content.
Nonlinearity & Aliasing
→Follow newly generated harmonics past Nyquist and examine the reflected frequencies that return as aliasing.
Oversampling — Move the Mirror
→Raise the internal sample rate to move the Nyquist boundary and reduce audible alias products from nonlinear processing.
Memoryless vs Stateful Nonlinearity
→Compare instantaneous waveshaping with nonlinear systems whose output also depends on their previous state.
Capture Lab — Modelling a Real Device
→Measure the behaviour of a real processor and turn observations into evidence for a model.
Modelling Strategies
→Compare circuit-derived, measurement/capture and learned approaches to reproducing the behaviour of audio systems.
Bit Crushing — Breaking the Digital Grid
→Reduce digital resolution deliberately and hear quantisation become a creative signal-processing effect.
Blocks, FFT & Spectral Processing
Move from sample-by-sample processing to blocks, transform those blocks into frequency data, and build toward spectral manipulation, reconstruction and vocoding.
Buffers, Blocks & Latency
→See why realtime audio is processed in chunks and how buffer size connects computation to latency.
FFT — From Time to Frequency
→Transform a block of samples into a frequency-domain representation and connect waveform structure to spectrum.
FFT Size, Bins & Resolution
→Explore how block length and sample rate determine bin spacing and the frequency resolution of an FFT.
Windows & Spectral Leakage
→Discover why finite FFT blocks cause spectral leakage and how window functions trade resolution for sidelobe suppression.
Overlap & Reconstruction
→See why windowed transform blocks overlap and how overlap-add reconstructs a continuous output signal.
Draw on the Spectrum
→Manipulate frequency-domain magnitudes directly and hear spectral changes return to the time domain.
Spectral Playground
→Experiment more freely with frequency-domain transformations and the sounds produced by manipulating spectral data.
Vocoder & Filter Banks
→Analyse one sound across frequency bands and use its spectral shape to control another.
BONUSSpectral Images
→Explore the spectrum as a visual and creative surface, linking image-like structures with frequency-domain sound.