Creative Harmonisation of Melodies via Learning and Blending of Ontologies

Conceptual Blending and Creativity

New concepts may be invented by ‘exploring’ previously unexplored regions of a given conceptual space (exploratory creativity) or transforming in novel ways established concepts (transformational creativity) or by making associations between different conceptual spaces that are not directly linked (combinational creativity) (Boden 2009).

Conceptual blending is a cognitive theory developed by Fauconnier and Turner (2003; Blending and Conceptual Integration webpage) whereby elements from diverse, but structurally-related, mental spaces are ‘blended’ giving rise to new conceptual spaces that often possess new powerful interpretative properties allowing better understanding of known concepts or the emergence of novel concepts altogether. It relates directly to Boden’s notion of combinational creativity.

In the context of the COINVENT project (Schorlemmer et al., 2014) a model is being developed that is based on Goguen's proposal of a Unified Concept Theory (Goguen, 2006), inspired by the category-theoretical formalisation of blending that employs the category-theoretical colimit operation to compute blends. This methodological framework incorporates important interdisciplinary research advances from cognitive science, artificial intelligence, formal methods and computational creativity.

Evaluating CHAMELEON - in the context of the (MIS 5005182) project

There is an ongoing projet for "Evaluating the contribution of the CHAMELEON system in human musical creativity" (MIS 5005182); more information can be found in this page. More information will be available after the results have been published: disclosing details to potential participants at this point, would jeopardize the integrity of possible complementary experiments that might be necessary in the near future. 

The CHAMELEON harmonisation assistant

As an illustration of the COINVENT model’s potential, a proof-of-concept computational creative system is developed that learns harmonies from diverse idioms, generates novel harmonisations of given melodies in the learned harmonic styles or in blended spaces of diverse harmonic idioms.

The CHAMELEON harmoniser is a melodic harmonisation assistant that is
•    adaptive: learns from data (Kaliakatsos-Papakostas et al., 2016a),
•    general: can cope with any tonal or non-tonal harmonic idiom (Cambouropoulos et al., 2014),
•    modular: learns different aspects of harmonic structure (Kaliakatsos-Papakostas et al., 2016) such as chord types (Cambouropoulos et al., 2014), chord transitions (Kaliakatsos-Papakostas and Cambouropoulos, 2014), cadences and voice-leading (Makris et al., 2015),
•    creative: not only generates novel harmonies for a given melody in a selected learned harmonic idiom (Kaliakatsos-Papakostas et al., 2016a), but may also blend different harmonies generating novel harmonic spaces altogether (Kaliakatsos-Papakostas et al., 2016b).

The various components of the proposed model are explained in a number of papers, and, examples of creative harmonisations of different melodies are presented:

•    Description of Harmonic Dataset for learning diverse harmonic styles
•    Examples of Input Melodies
•    Examples of melodic harmonisations in diverse harmonic idioms
•    Examples of melodic harmonisation based of blended harmonic spaces
•    Examples of creative use of the CHAMELEON assistant by composers


Boden, M. A. (2009). Computer Models of Creativity. AI Magazine, 30(3):23.

Cambouropoulos, E., Kaliakatsos-Papakostas, M., and Tsougras, C. (2014). An idiom-independent representation of chords for computational music analysis and generation. In Proceeding of the joint 11th Sound and Music Computing Conference (SMC) and 40th International Computer Music Conference (ICMC), ICMC–SMC 2014.

Fauconnier, G. and Turner, M. (2003). The Way We Think: Conceptual Blending And The Mind’s Hidden Complexities. Basic Books, New York, reprint edition.

Goguen, J. (2006). Mathematical Models of Cognitive Space and Time. In D. Andler, Y. Ogawa, M. Okada, and S. Watanabe (eds.), Reasoning and Cognition, Interdisciplinary Conference Series on Reasoning Studies, volume 2. Keio University Press.

Kaliakatsos-Papakostas, M. and Cambouropoulos, E. (2014). Probabilistic harmonisation with fixed intermediate chord constraints. In Proceeding of the joint 11th Sound and Music Computing Conference (SMC) and 40th International Computer Music Conference (ICMC), ICMC–SMC 2014.

Kaliakatsos-Papakostas, M., Makris, D., Tsougras, C., and Cambouropoulos, E. (2016a). Learning and creating novel harmonies in diverse musical idioms: An adaptive modular melodic harmonisation system. Journal of Creative Music Systems, 1(1).

Kaliakatsos-Papakostas, M., Queiroz, M., Tsougras, C., and Cambouropoulos, E. (2016b). Conceptual blending of harmonic spaces for creating melodic harmonisation. Journal of New Music Research:(Submitted).

Makris, D., Kaliakatsos-Papakostas, M., and Cambouropoulos, E. (2015). A probabilistic approach to determining bass voice leading in melodic harmonisation. In Mathematics and Computation in Music: Proceedings of the 5th International Conference, MCM 2015, London, UK. Springer, Berlin.

Schorlemmer, M., Smaill, A., Ku ̈hnberger, K.-U., Kutz, O., Colton, S., Cambouropoulos, E., and Pease, A. (2014). Coinvent: Towards a computational concept invention theory. In 5th Interna- tional Conference on Computational Creativity (ICCC) 2014.

CCM  group
CCM group website
AUTH website
COINVENT website