Tag: AI music

The Continuator: Musical Interaction With Style (Pachet, 2003)

The Continuator: Musical Interaction With Style

Author: François Pachet

Reference: Journal of New Music Research,
Vol. 32, No. 3, pp. 333–341, 2003.

Resources

Abstract

The Continuator is an interactive musical system designed to generate stylistically consistent
continuations of a musician’s playing in real time. It bridges the gap between traditional
interactive systems, which often lack stylistic coherence, and style imitation systems, which are
usually not interactive. The system learns statistical models of a performer’s style from
example material based on variable-order Markov chains, and uses them to generate responses
that follow the same stylistic patterns while remaining reactive and improvisatory. Experiments
with both professional musicians and children show that the Continuator can be perceived as a
meaningful musical partner, capable of extending the player’s ideas without merely copying them.

Highlights

  • Introduces the Continuator, a real-time interactive system that imitates and extends a musician’s style.
  • Uses variable-order Markov models to capture stylistic features such as rhythm, phrasing and harmony.
  • Bridges two worlds: style imitation systems and interactive performance systems.
  • Includes qualitative experiments with professional musicians and children, showing strong engagement and sense of “musical dialogue”.
  • A foundational paper in machine improvisation and style modelling; widely cited in computer music and AI creativity research.

Keywords

Interactive music systems; machine improvisation; style modelling; variable-order Markov models;
musical interaction; AI music; human–computer interaction.

BibTeX

@article{pachet2003continuator,
  title   = {The Continuator: Musical Interaction With Style},
  author  = {Pachet, Fran{\c{c}}ois},
  journal = {Journal of New Music Research},
  volume  = {32},
  number  = {3},
  pages   = {333--341},
  year    = {2003}
}

Why this paper matters

This paper is one of the earliest and most influential demonstrations of machine
improvisation in a real performance context. The Continuator showed that a computer can learn
a performer’s style and respond in real time in a way that feels both coherent and surprising,
effectively acting as a “virtual musician”. It laid the conceptual and technical groundwork for
later systems in interactive AI music and remains a reference for research on stylistic
imitation and human–machine musical dialogue.

DeepBach: A Steerable Model for Bach Chorales Generation

DeepBach: A Steerable Model for Bach Chorales Generation

Authors: Gaëtan Hadjeres, François Pachet, Frank Nielsen

Reference: Proceedings of the 34th International Conference on Machine Learning (ICML),
pp. 1362–1371, 2017.

Resources

Abstract

DeepBach introduces a generative model capable of producing Bach-style four-part chorales with
fine-grained user control. The system combines a pseudo-Gibbs sampling scheme with deep neural
architectures trained on the full Bach chorale corpus. Unlike earlier models, DeepBach allows
users to interactively constrain musical structure—such as meter, chord progressions, or specific
voices—while maintaining stylistic coherence. Human evaluation experiments show that many
DeepBach harmonizations are indistinguishable from genuine Bach chorales, making it one of the
first successful demonstrations of steerable symbolic music generation.

Highlights

  • Deep generative model for four-part Bach-style chorales using pseudo-Gibbs sampling.
  • Fully steerable: users can fix notes, chords, cadences, meter, and specific voices.
  • Generated chorales respect voice-leading and tonal grammar learned from the Bach corpus.
  • Human listening tests show many harmonizations are indistinguishable from genuine Bach.
  • One of the most cited works in neural symbolic music generation and AI-assisted composition.

Keywords

Deep learning for music; symbolic music generation; steerable generative models; Bach chorales;
pseudo-Gibbs sampling; computational creativity; music information retrieval.

BibTeX

@inproceedings{hadjeres2017deepbach,
  title        = {DeepBach: a Steerable Model for Bach Chorales Generation},
  author       = {Hadjeres, Ga{\"e}tan and Pachet, Fran{\c{c}}ois and Nielsen, Frank},
  booktitle    = {Proceedings of the 34th International Conference on Machine Learning},
  pages        = {1362--1371},
  year         = {2017}
}

Why this paper matters

DeepBach was one of the first systems to show that deep learning could convincingly generate music
in a highly codified style, to the point of fooling musicians in blind tests. It also framed
symbolic music generation as an interactive and steerable process, where human
users retain control over musical structure while the model fills in the details. This idea has
influenced many later AI music systems and remains a reference point for controllable music models.

A Taxonomy of Musical Genre (Pachet & Cazaly, RIAO 2000)

A Taxonomy of Musical Genre

François Pachet & Daniel Cazaly — RIAO 2000, Paris

Abstract

This paper proposes a structured approach to classifying musical genres based on metadata consistency, hierarchical structure, and content-based similarities. It highlights the challenges of defining genre boundaries, the lack of standardised taxonomies in the music industry, and the implications for music retrieval systems.

Download

Citation (BibTeX)

@inproceedings{pachet2000taxonomy,
  title={A Taxonomy of Musical Genre},
  author={François Pachet and Daniel Cazaly},
  booktitle={RIAO 2000 -- Content-Based Multimedia Information Access},
  address={Paris, France},
  year={2000},
  url={https://www.francoispachet.fr/wp-content/uploads/2021/01/pachet-00-RIAO.pdf}
}