Towards the generation of musical explanations with GPT-3

Stephen James Krol; Maria Teresa Llano; Jon McCormack

doi:10.1007/978-3-031-03789-4_9

Towards the generation of musical explanations with GPT-3

Stephen James Krol, Maria Teresa Llano, Jon McCormack

Department of Human Centred Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

2 Citations (Scopus)

Abstract

Open AI’s language model, GPT-3, has shown great potential for many NLP tasks, with applications in many different domains. In this work we carry out a first study on GPT-3’s capability to communicate musical decisions through textual explanations when prompted with a textual representation of a piece of music. Enabling a dialogue in human-AI music partnerships is an important step towards more engaging and creative human-AI interactions. Our results show that GPT-3 lacks the necessary intelligence to really understand musical decisions. A major barrier to reach a better performance is the lack of data that includes explanations of the creative process carried out by artists for musical pieces. We believe such a resource would aid the understanding and collaboration with AI music systems.

Original language	English
Title of host publication	11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings
Editors	Tiago Martins, Nereida Rodríguez-Fernández, Sérgio M. Rebelo
Place of Publication	Cham Switzerland
Publisher	Springer
Pages	131-147
Number of pages	17
ISBN (Electronic)	9783031037894
ISBN (Print)	9783031037887
DOIs	https://doi.org/10.1007/978-3-031-03789-4_9
Publication status	Published - 2022
Event	International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2022 - Madrid, Spain Duration: 20 Apr 2022 → 22 Apr 2022 Conference number: 11th https://link.springer.com/book/10.1007/978-3-031-03789-4 (Proceedings) https://www.evostar.org/2022/evomusart/ (Website)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	13221
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2022
Abbreviated title	EvoMUSART 2022
Country/Territory	Spain
City	Madrid
Period	20/04/22 → 22/04/22
Internet address	https://link.springer.com/book/10.1007/978-3-031-03789-4 (Proceedings) https://www.evostar.org/2022/evomusart/ (Website)

Keywords

Explainability
GPT3
Music

Access to Document

10.1007/978-3-031-03789-4_9

Cite this

Krol, S. J., Llano, M. T., & McCormack, J. (2022). Towards the generation of musical explanations with GPT-3. In T. Martins, N. Rodríguez-Fernández, & S. M. Rebelo (Eds.), 11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings (pp. 131-147). (Lecture Notes in Computer Science; Vol. 13221). Springer. https://doi.org/10.1007/978-3-031-03789-4_9

Krol, Stephen James ; Llano, Maria Teresa ; McCormack, Jon. / Towards the generation of musical explanations with GPT-3. 11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings. editor / Tiago Martins ; Nereida Rodríguez-Fernández ; Sérgio M. Rebelo. Cham Switzerland : Springer, 2022. pp. 131-147 (Lecture Notes in Computer Science).

@inproceedings{f65e3199335340cd9e705ff92b4c6272,

title = "Towards the generation of musical explanations with GPT-3",

abstract = "Open AI{\textquoteright}s language model, GPT-3, has shown great potential for many NLP tasks, with applications in many different domains. In this work we carry out a first study on GPT-3{\textquoteright}s capability to communicate musical decisions through textual explanations when prompted with a textual representation of a piece of music. Enabling a dialogue in human-AI music partnerships is an important step towards more engaging and creative human-AI interactions. Our results show that GPT-3 lacks the necessary intelligence to really understand musical decisions. A major barrier to reach a better performance is the lack of data that includes explanations of the creative process carried out by artists for musical pieces. We believe such a resource would aid the understanding and collaboration with AI music systems.",

keywords = "Explainability, GPT3, Music",

author = "Krol, {Stephen James} and Llano, {Maria Teresa} and Jon McCormack",

note = "Funding Information: Acknowledgements. The work presented here was funded by an Early Career Researcher Seed grant awarded by the Faculty of IT at Monash University. Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.; International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2022, EvoMUSART 2022 ; Conference date: 20-04-2022 Through 22-04-2022",

year = "2022",

doi = "10.1007/978-3-031-03789-4_9",

language = "English",

isbn = "9783031037887",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "131--147",

editor = "Tiago Martins and Nereida Rodr{\'i}guez-Fern{\'a}ndez and Rebelo, {S{\'e}rgio M.}",

booktitle = "11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings",

url = "https://link.springer.com/book/10.1007/978-3-031-03789-4, https://www.evostar.org/2022/evomusart/",

}

Krol, SJ, Llano, MT & McCormack, J 2022, Towards the generation of musical explanations with GPT-3. in T Martins, N Rodríguez-Fernández & SM Rebelo (eds), 11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings. Lecture Notes in Computer Science, vol. 13221, Springer, Cham Switzerland, pp. 131-147, International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2022, Madrid, Spain, 20/04/22. https://doi.org/10.1007/978-3-031-03789-4_9

Towards the generation of musical explanations with GPT-3. / Krol, Stephen James; Llano, Maria Teresa; McCormack, Jon.
11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings. ed. / Tiago Martins; Nereida Rodríguez-Fernández; Sérgio M. Rebelo. Cham Switzerland: Springer, 2022. p. 131-147 (Lecture Notes in Computer Science; Vol. 13221).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Towards the generation of musical explanations with GPT-3

AU - Krol, Stephen James

AU - Llano, Maria Teresa

AU - McCormack, Jon

N1 - Conference code: 11th

PY - 2022

Y1 - 2022

N2 - Open AI’s language model, GPT-3, has shown great potential for many NLP tasks, with applications in many different domains. In this work we carry out a first study on GPT-3’s capability to communicate musical decisions through textual explanations when prompted with a textual representation of a piece of music. Enabling a dialogue in human-AI music partnerships is an important step towards more engaging and creative human-AI interactions. Our results show that GPT-3 lacks the necessary intelligence to really understand musical decisions. A major barrier to reach a better performance is the lack of data that includes explanations of the creative process carried out by artists for musical pieces. We believe such a resource would aid the understanding and collaboration with AI music systems.

AB - Open AI’s language model, GPT-3, has shown great potential for many NLP tasks, with applications in many different domains. In this work we carry out a first study on GPT-3’s capability to communicate musical decisions through textual explanations when prompted with a textual representation of a piece of music. Enabling a dialogue in human-AI music partnerships is an important step towards more engaging and creative human-AI interactions. Our results show that GPT-3 lacks the necessary intelligence to really understand musical decisions. A major barrier to reach a better performance is the lack of data that includes explanations of the creative process carried out by artists for musical pieces. We believe such a resource would aid the understanding and collaboration with AI music systems.

KW - Explainability

KW - GPT3

KW - Music

UR - http://www.scopus.com/inward/record.url?scp=85128907121&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-03789-4_9

DO - 10.1007/978-3-031-03789-4_9

M3 - Conference Paper

AN - SCOPUS:85128907121

SN - 9783031037887

T3 - Lecture Notes in Computer Science

SP - 131

EP - 147

BT - 11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings

A2 - Martins, Tiago

A2 - Rodríguez-Fernández, Nereida

A2 - Rebelo, Sérgio M.

PB - Springer

CY - Cham Switzerland

T2 - International Conference on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2022

Y2 - 20 April 2022 through 22 April 2022

ER -

Krol SJ, Llano MT, McCormack J. Towards the generation of musical explanations with GPT-3. In Martins T, Rodríguez-Fernández N, Rebelo SM, editors, 11th International Conference, EvoMUSART 2022 Held as Part of EvoStar 2022 Madrid, Spain, April 20–22, 2022 Proceedings. Cham Switzerland: Springer. 2022. p. 131-147. (Lecture Notes in Computer Science). doi: 10.1007/978-3-031-03789-4_9