Home - mtpearce/idyom GitHub Wiki

IDyOM (Information Dynamics of Music) is a system for constructing multiple-viewpoint, variable-order Markov models for predictive modelling of probabilistic structure in symbolic, sequential auditory domains such as music. IDyOM acquires knowledge about a domain through statistical learning and generates conditional probability distributions representing the estimated likelihood of each event in a sequence, plus associated information-theoretic measures, given the preceding context and prior short- and long-term training of the model. The code contains modules for music representation, predicting phrase boundaries, simulating perception of similarity and generation of auditory sequences.

For more information, see Pearce (2005), Pearce (2025) and other related publications.

This wiki provides documentation for IDyOM users and developers. The IDyOM documentation assumes a basic knowledge of Common Lisp, and familiarity with the system's purpose and underlying theory.

See also the tutorial: https://github.com/mtpearce/idyom-tutorial

User documentation

• Installing and loading IDyOM and its prerequisites

• Database management

• Multiple viewpoints

  • Working with viewpoints
  • List of viewpoints

• Running IDyOM

  • IDyOM Parameters
  • IDyOM Output
  • IDyOM Examples

• Applying IDyOM beyond prediction

  • Segmentation
  • Similarity
  • Generation
  • Key‐finding

• Related software

• Troubleshooting

Developer documentation

This documentation is designed to help with the development of IDyOM.

• Code Architecture
• Development roadmap
• Specific modules:
  • Adding new viewpoints
  • The ppm-star package
  • The resampling package
  • The viewpoint-selection package