Chaotic Function Generator — User Guide

Mathematical chaos synthesis: generates complex, non-repeating waveforms from deterministic yet unpredictable functions for experimental sound design and algorithmic composition.

Author: Shai Cohen Affiliation: Department of Music, Bar-Ilan University, Israel Version: 0.1 (2025) License: MIT License Repo: https://github.com/ShaiCohen-ops/Praat-plugin_AudioTools

Contents:

What this does Quick start Chaos Theory Function Library Parameters Applications

What this does

This script implements a Chaotic Function Generator — a sophisticated mathematical tool that synthesizes audio from deterministic yet unpredictable mathematical functions. By exploiting the sensitive dependence on initial conditions characteristic of chaotic systems, the generator creates complex, non-repeating waveforms that are mathematically precise yet perceptually rich and evolving. The system includes 20 pre-defined chaotic functions plus custom formula support.

Key Features:

20 Chaotic Functions — Pre-defined mathematical systems with distinct characters
Custom Formula Support — User-defined mathematical expressions
Real-time Visualization — Graphical display of generated waveforms
Automatic Peak Normalization — Ensures optimal playback levels
Division-by-Zero Protection — Mathematical stability for all functions
Deterministic Chaos — Reproducible yet complex results
Educational Tool — Hear mathematical concepts as sound

What is chaotic synthesis? Traditional synthesis methods (subtractive, additive, FM) create predictable, periodic waveforms. Chaotic synthesis uses mathematical functions that exhibit chaos theory properties: deterministic equations that produce unpredictable, complex behavior due to extreme sensitivity to initial conditions. These functions create sounds that are neither purely random (like noise) nor purely periodic (like sine waves), but exist in a rich middle ground of structured complexity. The sounds evolve endlessly without repeating, making them ideal for creating organic, living textures.

Technical Implementation: (1) Function Selection: User chooses from 20 pre-defined chaotic systems or enters custom formula. (2) Mathematical Safety: Automatic epsilon values prevent division by zero at domain boundaries. (3) Waveform Generation: Praat's formula interpreter evaluates the mathematical expression across the time domain. (4) Visualization: The generated waveform is displayed graphically for immediate feedback. (5) Audio Normalization: Peak amplitude scaled to 0.99 for optimal playback. (6) Playback: Resulting sound played through audio output. The core innovation is the translation of abstract mathematical chaos into audible, musically useful sounds while maintaining mathematical integrity and computational stability.

Quick start

In Praat, ensure no objects are selected (generates new sound).
Run script… → chaotic_function_generator.praat.
Choose function from the 20 pre-defined options or select "Custom…".
Set duration_seconds (default 1.0) for output length.
Set sampling_rate_Hz (default 44100) for audio quality.
For custom functions, enter custom_formula using Praat syntax.
Click OK — sound generated, visualized, and played automatically.
Experiment with different functions to discover their unique sonic characters.

Quick tip: Start with sin(1/x) for classic chaotic tones, Logistic Map for mathematical chaos, or exp(-1/x^2)*sin(50*x) for complex textures. Use shorter durations (0.5-2.0 seconds) for quick experimentation. The visualization helps understand the waveform structure — chaotic functions typically show complex, non-repeating patterns. All functions automatically include small epsilon values to prevent mathematical errors at x=0 and x=1, ensuring stable operation.

Important: MATHEMATICAL INTENSITY — some functions are computationally expensive and may cause Praat to become unresponsive with very long durations (>10 seconds). Custom formulas must use valid Praat mathematical syntax. The domain is automatically normalized to [0,1] for time variable x. Very high sampling rates (>96 kHz) with complex functions may slow generation. The sounds are synthesized from scratch — no input audio is used or modified. Some functions may produce very quiet results that require amplification after generation. The system is deterministic — same parameters always produce identical output.

Chaos Theory Fundamentals

What Makes a System Chaotic?

Key Properties of Chaos

Mathematical definition:

Chaotic systems exhibit: 1. Deterministic: Follow precise mathematical rules 2. Sensitive dependence: Small changes in initial conditions → large differences in outcome 3. Topological mixing: Eventually explores all possible states 4. Dense periodic orbits: Infinite number of unstable periodic solutions In audio terms: - Same function always produces same waveform from same start - But waveform is complex and non-repeating - Creates "structured randomness" perceptually - Mathematical precision with perceptual richness

Why 1/x Creates Chaos

The singularity principle:

Rapid oscillation: As x→0, 1/x→∞ causing infinite frequency
Non-periodicity: No repeating pattern due to continuous change
Scale invariance: Similar patterns at different time scales
Mathematical instability: Extreme sensitivity near x=0

Domain and Stability

Time Domain Mapping

Normalized time variable:

Time variable: x ranges from 0 to 1 over duration x = time / duration This means: Start: x = 0 (beginning of sound) End: x = 1 (end of sound) Critical points: x = 0: Beginning, potential mathematical singularities x = 1: End, some functions have special behavior Epsilon protection: x → x + ε where ε = 1e-4 Prevents division by zero at boundaries

Mathematical Safety System

Automatic singularity avoidance:

Original formula: sin(1/x) Protected formula: sin(1/(x + 1e-4)) Original formula: sin(1/(x*(1-x))) Protected formula: sin(1/((x+1e-4)*((1-x)+1e-4))) This ensures: - No division by zero errors - Stable computation across entire domain - Preservation of chaotic character - Smooth operation in Praat environment

Chaos vs Randomness

Key Differences

Mathematical distinction:

🎲 Random Noise

Characteristics:

Truly unpredictable
No underlying pattern
Maximum entropy
No memory of previous states

Audio example: White noise

🌀 Chaotic System

Characteristics:

Deterministic but unpredictable
Complex underlying structure
Structured complexity
Complete memory (dependent on initial conditions)

Audio example: sin(1/x) synthesis

Classical Chaotic Systems

Logistic Map

Mathematical definition:

Logistic Map equation: xₙ₊₁ = r · xₙ · (1 - xₙ) Where: xₙ = value at step n (between 0 and 1) r = growth rate parameter In this implementation: r = 3.9 (chaotic regime) Formula: 3.9*x*(1-x) Behavior: - For r < 3.57: periodic behavior - For r ≥ 3.57: chaotic behavior - At r = 3.9: fully developed chaos

Tent Map

Piecewise linear chaos:

Tent Map definition: xₙ₊₁ = μ · min(xₙ, 1 - xₙ) In this implementation: μ = 2 (maximal chaos) Formula: if x<0.5 then 2*x else 2*(1-x) endif Properties: - Piecewise linear function - Uniform invariant distribution - Exact solutions possible - Simpler than logistic map but equally chaotic

📈 Visualizing Chaos

sin(1/x) near x=0:

As x approaches 0, 1/x approaches infinity

sin(1/x) oscillates infinitely rapidly

Creates dense, non-repeating waveform

Logistic Map bifurcation:

At r=3.9, system never settles to fixed point

Continually visits different values in [0,1]

Creates unpredictable yet deterministic sequence

Function Library

Singularity-Based Functions

Function	Formula	Sonic Character
sin(1/x)	sin(1/(x+ε))	Classic chaotic oscillation, infinite frequency at start
*sin((1/x)(1/(1-x)))**	sin((1/(x+ε))*(1/((1-x)+ε)))	Dual singularity, complex both ends
*sin(1/(x(1-x)))**	sin(1/((x+ε)*((1-x)+ε)))	Product singularities, rich texture
*atan(1/x)sin(10/x)**	arctan(1/(x+ε))*sin(10/(x+ε))	Bounded chaos, smoother evolution

Multi-component Functions

Function	Formula	Sonic Character
Multi-sine/cosine	2sin(3/x)+3cos(5/x)+4*sin(6/x)+cos(3/x)	Rich harmonic chaos, complex beating
*sin(1/x) + 2sin(1/(1-x))**	sin(1/(x+ε)) + 2*sin(1/((1-x)+ε))	Mirrored chaos, symmetric structure
*sin(3/x)sin(5/(1-x))**	sin(3/(x+ε))*sin(5/((1-x)+ε))	Modulated chaos, product complexity
*tan(1/x)cos(1/(1-x))**	tan(1/(x+ε))*cos(1/((1-x)+ε))	Unbounded growth with modulation

Power Law Functions

Function	Formula	Sonic Character
*sin(1/x²)cos(1/(1-x)²)**	sin(1/(x+ε)²)*cos(1/((1-x)+ε)²)	Rapid divergence, extreme chaos
*sin(1/x)cos(1/x²)**	sin(1/(x+ε))*cos(1/(x+ε)²)	Multi-scale chaos, nested patterns
(sin(1/x)+sin(1/x³))/2	(sin(1/(x+ε))+sin(1/(x+ε)³))/2	Averaged chaos, smoothed result
cos(1/x³)+sin(1/(1-x)³)	cos(1/(x+ε)³)+sin(1/((1-x)+ε)³)	Cubic singularity, very rapid oscillation

Exponential and Logarithmic

Function	Formula	Sonic Character
*exp(-1/x²)sin(50x)*	exp(-1/(x+ε)²)sin(50x)	Gaussian-weighted oscillation
*sin(ln(x+0.01))cos(ln(1.01-x))**	sin(ln(x+0.01))*cos(ln(1.01-x))	Logarithmic stretching, slow evolution
*exp(sin(1/x))cos(1/(1-x))**	exp(sin(1/(x+ε)))*cos(1/((1-x)+ε))	Exponential modulation, growing complexity
*sin(1/x)sin(1/x²)sin(1/x³)*	sin(1/(x+ε))sin(1/(x+ε)²)sin(1/(x+ε)³)	Triple product, maximal complexity

Classical Chaotic Maps

Function	Formula	Sonic Character
Logistic Map	3.9x(1-x)	Mathematical chaos, unpredictable yet deterministic
Tent Map	min(2x, 2(1-x))	Piecewise linear chaos, sharp transitions

Specialized Functions

Function	Formula	Sonic Character
Weighted oscillation	(1/x)sin(10x)+(1/(1-x))cos(10(1-x))	Amplitude-modulated chaos
Square root chaos	sin(1/√(x+0.001))*cos(1/√(1.001-x))	Slower divergence, smoother chaos

Function Character Guide

🎵 Musical Chaos

For rich, evolving tones:

Multi-sine/cosine: Complex harmonic beating
exp(-1/x²)*sin(50*x): Musical oscillation with chaos
sin(1/x) + 2*sin(1/(1-x)): Balanced chaotic structure

🌊 Textural Chaos

For complex textures:

sin(1/(x*(1-x))): Rich, dense patterns
sin(1/x)*sin(1/x²)*sin(1/x³): Maximum complexity
Logistic Map: Mathematical texture

⚡ Extreme Chaos

For rapid, intense patterns:

sin(1/x²)*cos(1/(1-x)²): Very rapid oscillation
cos(1/x³)+sin(1/(1-x)³): Cubic singularity intensity
Tent Map: Sharp, discontinuous chaos

Parameters

Core Generation Parameters

Parameter	Type	Default	Description
function	optionmenu	sin(1/x)	Mathematical function for synthesis (20 options + custom)
duration_seconds	real	1.0	Length of generated sound in seconds
sampling_rate_Hz	integer	44100	Audio sampling rate (samples per second)
custom_formula	text	sin(1/x)	User-defined mathematical expression (Praat syntax)

Automatic Processing Parameters

Parameter	Behavior	Purpose
Epsilon protection	Automatic	Prevents division by zero (ε=1e-4)
Peak normalization	Automatic	Scales output to 0.99 peak amplitude
Waveform visualization	Automatic	Draws generated waveform for inspection
Auto-playback	Automatic	Plays sound immediately after generation

Parameter Guidelines

⏱️ Duration Settings

Short (0.1-0.5s): Quick experimentation, impulse-like sounds

Medium (0.5-2.0s): General use, manageable file sizes

Long (2.0-10.0s): Extended textures, may slow generation

Very Long (>10.0s): Risk of Praat slowdown, complex functions

🎚️ Sampling Rate Settings

CD Quality (44100 Hz): Standard audio, good performance

High Quality (48000-96000 Hz): Extended frequency range

Low Quality (22050 Hz): Faster generation, lo-fi character

Very High (>96000 Hz): May slow complex functions

Applications

Sound Design and Synthesis

Use case: Creating unique, evolving synthetic textures

Technique: Generate chaotic sounds, then process with filters/effects

Example: Use Logistic Map as source for granular synthesis

Algorithmic Composition

Use case: Generating musical material with mathematical structure

Technique: Use chaotic functions to create pitch/rhythm sequences

Example: Extract features from chaos for melody generation

Scientific and Educational Use

Use case: Demonstrating mathematical concepts through sound

Technique: Compare different chaotic systems auditorily

Example: Hear the difference between periodic and chaotic behavior

Experimental Music

Use case: Creating non-repeating, organic musical structures

Technique: Layer multiple chaotic generators

Example: Create ever-changing drone textures

Practical Workflow Examples

🎹 Chaotic Melodic Source

Goal: Create evolving melodic patterns

Settings:

Function: exp(-1/x²)*sin(50*x)
Duration: 5.0 seconds
Sampling Rate: 44100 Hz

Process: Generate, then apply pitch shifting and time stretching

Result: Musical chaos with pitched characteristics

🌫️ Textural Bed Creation

Goal: Generate complex background textures

Settings:

Function: sin(1/(x*(1-x)))
Duration: 10.0 seconds
Sampling Rate: 48000 Hz

Process: Generate multiple layers, apply reverb and filtering

Result: Rich, non-repeating textural background

🔬 Mathematical Demonstration

Goal: Compare different chaotic systems

Settings:

Functions: sin(1/x), Logistic Map, Tent Map
Duration: 2.0 seconds each
Sampling Rate: 44100 Hz

Process: Generate and compare waveforms and sounds

Result: Audible differences between chaotic systems

Advanced Techniques

Custom formula creation:

Use x as time variable: Ranges from 0 to 1 over duration
Praat math functions: sin, cos, tan, exp, ln, sqrt, arctan, etc.
Conditional logic: if...then...else...endif statements
Multiple operations: Combine functions for complexity
Avoid extreme values: Very large outputs may cause clipping
Test with visualization: Use drawing to verify function behavior

Post-processing strategies:

Filtering: Shape frequency content of chaotic sounds
Time-stretching: Extend chaotic evolution
Layering: Combine multiple chaotic generators
Modulation: Use chaos to modulate other sounds
Granular processing: Break chaos into micro-sounds
Spatialization: Place chaotic elements in stereo field

Troubleshooting Common Issues

Problem: Praat becomes unresponsive during generation
Cause: Very long duration with complex function
Solution: Use shorter durations (1-5 seconds), simpler functions

Problem: Generated sound is very quiet
Cause: Function produces small output values
Solution: Use Amplify... function in Praat to increase gain

Problem: Custom formula produces errors
Cause: Invalid Praat syntax or mathematical errors
Solution: Check formula syntax, avoid division by zero, test simple first

Problem: Sound is mostly noise with little structure
Cause: Extreme chaotic function or inappropriate parameters
Solution: Try different functions, adjust duration, use visualization

Problem: No sound generated or played
Cause: Audio system issues or extreme function values
Solution: Check Praat audio settings, try basic function first