Development of Adaptive Systems for the Production of Mixed Music: Machine Learning as a Search Tool in the Context of Human-Machine Interaction.

• Grant number: 13/17329-5
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: November 01, 2013
• End date: April 30, 2015
• Field of knowledge: Linguistics, Literature and Arts - Arts - Music
• Principal Investigator: Jônatas Manzolli
• Grantee: Tiago Fernandes Tavares
• Host Institution: Núcleo Interdisciplinar de Comunicação Sonora (NICS). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

This project regards the construction of an exploration environment for tools based on machine listening. More precisely, it focuses on the study and development of a new knowledge area related to Music Information Retrieval (MIR) with computational support. Essentially, this tool uses MIR algorithms, which are based on digital signal processing and machine learning. The proposed study aims at music composition processes related to mixed pieces with real time processing. In this kind of musical process, there in an inherent need for the identification and grouping of sonic characteristics that can be used as material for the organization of the temporal interaction of a piece. This task is frequently considered as part of the creative process, but the musician can be helped by a computational tool capable of identifying auditory similarities. The tool would be able to operate in a large dataset, granting access to an amount of data that cannot feasibly be analyzed manually.Also, it is necessary to develop interfaces so that a musician can usethe tool and understand the provided information. This should accountfor the individuality of a creative process, that is, the need for thepersonalization of the workspace. Hence, there is great relevance inthe interaction of thecandidate with researchers and musicians from NICS (InterdisciplinaryNucleus for Sound Communication), place where the project will be developed.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

TAVARES, TIAGO FERNANDES. An interactive audio-visual installation using ubiquitous hardware and web-based software deployment. PEERJ COMPUTER SCIENCE, MAY 2015. (13/17329-5)

TAVARES, TIAGO FERNANDES. An interactive audio-visual installation using ubiquitous hardware and web-based software deployment. PEERJ COMPUTER SCIENCE, v. N/A, p. 15-pg., 2015-05-01. (13/17329-5)