Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Recurso digital |
| Language: | |
| Published: |
Zenodo
2026
|
| Online Access: | https://doi.org/10.5281/zenodo.19417545 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866901890775646208 |
|---|---|
| author | Madaghiele, Vincenzo Fasciani, Stefano Erdem, Çağrı |
| author_facet | Madaghiele, Vincenzo Fasciani, Stefano Erdem, Çağrı |
| contents | <p>This repository contains video examples, and software configurations supporting the results of the user studies included in the article: </p> <p>"Personalizing Behaviors of the Multi-Agent Autonomous Looper with a Corpus of Annotated Co-performances". Vincenzo Madaghiele, Stefano Fasciani, Çagri Erdem. Submitted to <em>Organised Sound, 2026.</em></p> <p><strong>Abstract:</strong></p> <p>This article introduces a corpus-based method to personalize the behavior of a music co-improvisation system that autonomously samples and loops segments of an improvised performance by integrating machine listening with a rule-based framework. The system, named Multi-Agent Autonomous Looper (MAAL), is personalized using a genetic algorithm that iteratively identifies optimal rules governing the agents’ behavior according to musicians’ personal style, preferences and aesthetic objectives. The corpus annotations represent the decisions that agents should learn to make in response to different musicians’ behaviors, reflecting the musicians’ styles, ideas, and aesthetic preferences. After detailing the design of the evolutionary search algorithm, we present the results of a study involving five expert improvising musicians. These musicians applied the proposed method to personalize the MAAL and co-improvised with it. We then assessed the extent to which the MAAL's behavior, trained on their personalized data, aligned with their individual aesthetic preferences. Findings show that personalization can improve musician-agents’ co-adaptation by establishing a defined aesthetic vocabulary of the co-creative system. Moreover, we found that automating the functions of sampling and layering loops shifts musicians’ focus towards responding and listening, while simultaneously allowing them to take on the role of a manager/curator of the overall aesthetic outcome.</p> <p> </p> <p>The videos were realized with the version of the MAAL software used for performing the user studies in the article. All participants voluntarily agreed to have their identities disclosed in the article, and their videos published as examples. </p> <p>The last updated version of the MAAL software is hosted at this Github repository: <a href="https://github.com/vincenzomadaghiele/genetic-MAAL">https://github.com/vincenzomadaghiele/genetic-MAAL.</a></p> <p>The MAAL is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19417545 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Personalizing Behaviors of the Multi-Agent Autonomous Looper with a Corpus of Annotated Co-performances - supplementary materials Madaghiele, Vincenzo Fasciani, Stefano Erdem, Çağrı <p>This repository contains video examples, and software configurations supporting the results of the user studies included in the article: </p> <p>"Personalizing Behaviors of the Multi-Agent Autonomous Looper with a Corpus of Annotated Co-performances". Vincenzo Madaghiele, Stefano Fasciani, Çagri Erdem. Submitted to <em>Organised Sound, 2026.</em></p> <p><strong>Abstract:</strong></p> <p>This article introduces a corpus-based method to personalize the behavior of a music co-improvisation system that autonomously samples and loops segments of an improvised performance by integrating machine listening with a rule-based framework. The system, named Multi-Agent Autonomous Looper (MAAL), is personalized using a genetic algorithm that iteratively identifies optimal rules governing the agents’ behavior according to musicians’ personal style, preferences and aesthetic objectives. The corpus annotations represent the decisions that agents should learn to make in response to different musicians’ behaviors, reflecting the musicians’ styles, ideas, and aesthetic preferences. After detailing the design of the evolutionary search algorithm, we present the results of a study involving five expert improvising musicians. These musicians applied the proposed method to personalize the MAAL and co-improvised with it. We then assessed the extent to which the MAAL's behavior, trained on their personalized data, aligned with their individual aesthetic preferences. Findings show that personalization can improve musician-agents’ co-adaptation by establishing a defined aesthetic vocabulary of the co-creative system. Moreover, we found that automating the functions of sampling and layering loops shifts musicians’ focus towards responding and listening, while simultaneously allowing them to take on the role of a manager/curator of the overall aesthetic outcome.</p> <p> </p> <p>The videos were realized with the version of the MAAL software used for performing the user studies in the article. All participants voluntarily agreed to have their identities disclosed in the article, and their videos published as examples. </p> <p>The last updated version of the MAAL software is hosted at this Github repository: <a href="https://github.com/vincenzomadaghiele/genetic-MAAL">https://github.com/vincenzomadaghiele/genetic-MAAL.</a></p> <p>The MAAL is free software: you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.</p> |
| title | Personalizing Behaviors of the Multi-Agent Autonomous Looper with a Corpus of Annotated Co-performances - supplementary materials |
| url | https://doi.org/10.5281/zenodo.19417545 |