Synthesis

Subtractive synthesis starts with harmonically rich waveforms and removes frequencies using filters. This intuitive approach mirrors how acoustic instruments work - starting with complex overtones and sculpting the timbre. The classic filter sweep and resonance effects are hallmarks of this synthesis method.

Additive synthesis builds sounds by combining multiple sine waves (harmonics) to create complex timbres. Each harmonic can be individually controlled for amplitude and frequency, allowing precise spectral sculpting. This method offers ultimate control over harmonic content but requires deep understanding of frequency relationships.

FM synthesis uses one oscillator to modulate the frequency of another, creating complex harmonics and metallic timbres. The modulating oscillator affects the carrier oscillator's pitch at audio rates, generating sidebands that create rich, inharmonic spectra perfect for bells, metallic sounds, and digital textures.

AM synthesis modulates the amplitude of a carrier signal with a modulator, creating tremolo effects and ring modulation. When the modulator frequency is in the audio range, it creates sum and difference frequencies, producing metallic, bell-like tones. AM is simpler than FM but can create equally dramatic effects.

Wavetable synthesis uses stored single-cycle waveforms that can be smoothly interpolated and modulated in real-time. By scanning through wavetables with different harmonic content, it creates evolving timbres impossible with traditional oscillators. Modern wavetable synths offer extensive modulation and real-time manipulation.

Granular synthesis breaks audio into tiny grains (1-100ms) that can be manipulated independently for pitch, timing, and spatialization. This technique excels at time-stretching, pitch-shifting, and creating complex textures from any source material. It's particularly effective for atmospheric pads and evolving soundscapes.

Phase Distortion synthesis manipulates the phase of sine waves to create complex harmonics without using traditional filters. By altering the phase relationships during wave generation, it produces rich timbres with unique character. This method was popularized by Casio's CZ series synthesizers.

Vector synthesis blends multiple oscillators or sound sources using a joystick or automation, creating smooth morphing between different timbres. The vector position determines the mix of four sources, allowing real-time transitions between completely different sounds within a single patch.

Physical modeling synthesis simulates the physical properties of real instruments using mathematical models. Instead of playing back samples, it calculates how strings vibrate, how air moves through tubes, or how membranes resonate. This creates realistic, expressive sounds that respond naturally to playing dynamics.

Spectral synthesis analyzes and manipulates the frequency spectrum of sounds directly, allowing precise control over individual frequency components and their evolution over time. This method enables impossible transformations like cross-synthesis, spectral filtering, and harmonic restructuring.

Neural synthesis uses artificial intelligence and machine learning to generate and manipulate audio. These systems can learn from vast datasets to create new sounds, perform intelligent transformations, or generate music autonomously. This represents the cutting edge of synthesis technology.

Concatenative synthesis creates new sounds by selecting and concatenating small segments from a large database of recorded sounds. The system analyzes target characteristics and finds matching segments in the database, creating new combinations that maintain the desired musical and timbral qualities.

Sample-based synthesis uses recorded audio as the foundation, manipulating playback speed, pitch, and applying filters and effects. Modern samplers offer advanced features like time-stretching, multisampling, and complex modulation. This method combines the realism of recordings with the flexibility of synthesis.

Resynthesis analyzes existing audio and recreates it using synthesis parameters, allowing manipulation impossible with the original. This technique can extract harmonic, noise, and transient components separately, enabling independent control over each element for creative sound design.

Hybrid synthesis combines multiple synthesis methods within a single instrument, such as subtractive filters applied to FM operators, or granular processing of wavetables. This approach leverages the strengths of different synthesis types to create rich, complex sounds impossible with any single method.

Formant synthesis focuses on recreating vocal characteristics by emphasizing specific frequency regions (formants) that give vowels their distinctive character. This method is particularly effective for vocal synthesis, talking instruments, and creating human-like textures from electronic sources.

Chaos synthesis uses chaotic mathematical functions to generate unpredictable yet structured sounds. These non-linear systems produce complex, evolving timbres that never exactly repeat, creating organic-sounding textures perfect for ambient music and sound design applications.

Feedback synthesis creates sounds by feeding the output of a system back into its input, creating complex resonances and self-oscillation. This technique can produce everything from subtle harmonic enhancement to aggressive digital distortion and chaotic noise textures.

Algorithmic synthesis uses computer algorithms and mathematical procedures to generate sounds and musical structures. This approach can create complex, evolving compositions and soundscapes that would be impossible to program manually, often incorporating rules, probability, and generative processes.

Cross synthesis applies the characteristics of one sound to another, creating hybrid timbres that combine elements from both sources. This includes vocoding, convolution, and spectral techniques that can make one instrument sound like it's being filtered through another.