Papers by Mauricio Alves Loureiro
This project was born with the idea of developing a voice assistant to facilitate access to the l... more This project was born with the idea of developing a voice assistant to facilitate access to the library services of the Complutense University of Madrid through a mobile device. To this end, technologies have been used natural language processing to ensure that the system recognizes the requests made by a user and be able to return an appropriate response in a limited context. All this by simulating the behaviour of a human being so that the system is friendly. Thus, in order to obtain the information of copies requested in the application, use is made of the online catalogue of the library of the UCM, and in addition, an own database is used that stores basic information of the libraries in order to answer questions about institutional data of these. Resources, such as executables, code and sripts of the project can be accessed at the following link
Este proyecto nace con la idea de desarrollar un asistente de voz para facilitar el acceso a los ... more Este proyecto nace con la idea de desarrollar un asistente de voz para facilitar el acceso a los servicios de la biblioteca de la Universidad Complutense de Madrid a traves de un dispositivo movil. Para ello, se han utilizado tecnologias de procesamiento de lenguaje natural para conseguir que el sistema reconozca las peticiones que un usuario le realiza y ser capaz de devolver una respuesta apropiada en un contexto acotado. Todo ello, simulando el comportamiento de un ser humano para que el sistema sea amigable. Asi pues, con el objetivo de obtener la informacion de ejemplares solicitados en la aplicacion, se hace uso del catalogo online de la biblioteca de la UCM. Ademas, se utiliza una base de datos propia que almacena informacion basica de las bibliotecas con el fin de responder a preguntas sobre datos institucionales de estas. Puede acceder a todos los recursos, asi como ejecutables, codigo y scripts del proyecto en el siguiente enlace https://github.com/NILGroup/TFG-1819-Biblio...
Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)
O CEGeME - Centro de Pesquisa em Gesto e Expressão Musical é filiado ao Programa de Pós-Gradua... more O CEGeME - Centro de Pesquisa em Gesto e Expressão Musical é filiado ao Programa de Pós-Graduação em Música da Universidade Federal de Minas Gerais (UFMG), sediado na Escola de Música de Belo Horizonte, Brasil, desde 2008 é dedicado à pesquisa empírica da performance musical, que partem de informação de conteúdo musical extraída de sinais de áudio e da posição espacial tridimensional dos músicos, registradas durante uma performance musical. Com o objetivo de estabelecer uma abordagem musicológica a diferentes aspectos da expressividade musical, investigamos relações causais entre a intenção expressiva dos músicos e a forma como manipulam o material acústico e como se movem durante a execução de uma peça musical. O grupo atraiu postulantes de estudos de diversas áreas de conhecimento e músicos profissionais instigados por questões específicas envolvidas na expressividade musical, tais como, estudos sobre o uso da segunda válvula do trombone baixo moderno...
Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)
Although computational models for note onset detection have improved drastically in the last deca... more Although computational models for note onset detection have improved drastically in the last decade, mainly due to the advances brought by the field of Deep Learning, such models have not been perfected yet. When dealing with specific data, like clarinet recordings, those models still produce a significant number of false positives and negatives. In this paper, we evaluate pre-trained onset detection models from the library madmom on a dataset composed of solo clarinet recordings, in particular, to investigate their performance on this kind of data. Moreover, we use the clarinet dataset to train the same neural network (CNN) employed in one of those models, to investigate whether training the model on this specific data leads to an improvement when dealing with clarinet recordings. The results obtained from the model trained strictly on clarinet data are considerably better than those from models trained on generic data.
Abstract. In order to represent the variety of sounds a musical instrument may produce, it is nec... more Abstract. In order to represent the variety of sounds a musical instrument may produce, it is necessary to find a model that can cope with sound features independent from its scale. In this work, several models of timbre characterization were applied to sample notes in several intensity levels across the whole extension of a clarinet. These models were based on amplitude and frequency time-varying curves of partials, which were measured by Discrete Fourier Transform. Principal Component Analysis techniques and Genetic Algorithms were used to define spectral sub-spaces capable of representing all tested sounds and of grouping them. The K-means clustering algorithm was used to infer timbre classes. Self-Organizing Maps lead to results similar to those obtained by PCA representation and K-means algorithms. Resumo. Para se representar a variedade de sons que um instrumento musical é capaz de produzir, é necessário se utilizar modelos que podem lidar com um conjunto de parâmetros, indepe...
Music Perception, 2018
This article seeks to unveil quantitative relations between the patterns of movement recurrence o... more This article seeks to unveil quantitative relations between the patterns of movement recurrence of a group of expert clarinetists and expressive sonic manipulations they employ during their performances. The main hypothesis is that the recurrent ancillary gestures of musicians are closely related to their sounded expressive intentions, and that the expressive content imposed by them according to the music structure is reflected in their movement patterns. To conduct this multimodal investigation of expressiveness in music, movement and audio analyses of several clarinet performances of excerpts in the classical repertoire are presented and discussed in conjunction. The results show strong correlations between the recurrence pattern of clarinetists’ ancillary movements and expressive manipulations of timing, timbre, and loudness associated with melodic phrasing and harmonic and dynamic transitions in the performed music excerpts.
(Austria) in 2008, aims at connecting students and researchers from the broad spectrum of music r... more (Austria) in 2008, aims at connecting students and researchers from the broad spectrum of music research fields to envisage the forthcoming generation of research in systematic musicology. The focus of this conference is on one of the major challenges for researchers in systematic musicology, namely, how to deal with the inter-disciplinary nature inherent to our own field. In order to promote this question, the committees have been engaged in promoting a conference that is marked by academic independence, a large interdisciplinary span and an international scope. The review committee, composed of advanced PhD students and post docs, was responsible for 93 anonymous reviews and for the proposition and selection of keynotes. These efforts lead to this volume of proceedings, which includes 18 papers (66% of acceptance) and 27 authors representing 17 countries. The diversity of areas is represented by topics as diverse as music cognition, music performance, cross-cultural studies, acoustics, psychoacoustics, and music and movement. On behalf of SysMus09, we wish to thank our keynote speakers, professors Ian Cross (University of Cambridge) and Marcelo Wanderley (McGill University), as well as professor Marcel Brass (Ghent University) for their contribution to our conference programme. We also wish to thank our programme chairs, senior advisors and Manuela Marin for the support and care about the concept of the conference.
Brazilian Symposium on Computer Music, 2005
Abstract. In order to represent the variety of sounds a musical instrument may produce, it is nec... more Abstract. In order to represent the variety of sounds a musical instrument may produce, it is necessary to find a model that can cope with sound features independent from its scale. In this work, several models of timbre characterization were applied to sample notes in several intensity levels across the whole extension of a clarinet. These models were based on amplitude and frequency time-varying curves of partials, which were measured by Discrete Fourier Transform. Principal Component Analysis techniques and Genetic Algorithms ...
Opus, 2012
A series of studies have shown that musicians accompany themselves better than others. This is kn... more A series of studies have shown that musicians accompany themselves better than others. This is known as the self/other effect in action recognition and simulation and has been demonstrated in other action-perception tasks like recognizing handwriting and the outcomes of dart throwing. In this study, patterns of synchronization and performance coupling were analyzed among professional clarinetists while performing in a simulated orchestral environment. GLMM models (Generalized Linear Mixed Model) applied to acoustic parameters related to the musician's expressive intentions demonstrate a higher degree of coupling in performances where musicians accompanied themselves compared to when they accompanied other musicians., as well as an improvement in performance when accompanying other musicians as experiments repeated, suggesting musicians were capable of learning synchronization and performance coupling.
Lecture Notes in Computer Science, 2021
Singers perform different vocal adaptations to sing both lyrical and popular music. The mechanism... more Singers perform different vocal adaptations to sing both lyrical and popular music. The mechanism M1 (modal register) is predominant, but a small portion uses the mechanism M2 (falsetto register). In the lyrical singing, it is the countertenor who makes this peculiar vocal emission. The present study analyzed through nasofiberscopic, glottic, and supraglottic adjustments performed by countertenors, as well as the presence and level of the singer’s formant. The tasks (two interval leaps - a minor sixth and a perfect fourth – and a crescendo in sustained note) were taken from an aria by Georg Friedrich Haendel. During the crescendo, a lowering of the larynx and pharyngeal widening were observed; pharyngeal constrictions and laryngeal elevation were observed in both leaps executions; no glottal chink or air leak were detected. Singer’s formant was identified on all subjects. Professional countertenors performed better than non-professional because they had a higher level of singer'...
Proceedings of the SMC Conferences, Sep 14, 2014
This study aims at investigating if movements executed by musicians during musical performances a... more This study aims at investigating if movements executed by musicians during musical performances are related to their expressive intentions. We tested the hypothesis by identifying information patterns on body movement data collected in performances of clarinet duos under different experimental conditions. First, we identified information patterns recurrent enough for the recognition of musicians who performed the excerpts. Then, we found evidence for a change in the "gestural signature" of the instrumentalist follower imposed by the leader, across the different performance conditions.
The Routledge Companion to Embodied Music Interaction, 2017
Na guitarra eletrica, o vibrato e o bending sao tecnicas que, de acordo com o senso comum, tem um... more Na guitarra eletrica, o vibrato e o bending sao tecnicas que, de acordo com o senso comum, tem um papel importante na assinatura do guitarrista. Esse trabalho visa investigar caracteristicas na execucao do bending e do vibrato que diferenciam um guitarrista do outro. Propomos um modelo para a categorizacao das performances composto de quatro estagios: (1) extracao de altura; (2) segmentacao de notas; (3) extracao de descritores de bending e vibrato; (4) analise e categorizacao de performances baseada na parametrizacao proposta. Para validar o modelo, analisamos um conjunto de performances de dois trechos, executadas por oito guitarristas de expertise distinta.
Artigo publicado no III SMCT, em 2008.
A individualidade em uma performance musical esta relacionada ao conjunto das caracteristicas com... more A individualidade em uma performance musical esta relacionada ao conjunto das caracteristicas comuns as diferentes performances realizadas pelo mesmo musico, a partir das quais e possivel para os ouvintes distinguir o individuo que a realizou. Neste texto, apresentamos alguns estudos importantes que investigam a individualidade musical na performance, e discutimos conceitos teoricos, estrategias metodologicas e resultados importantes ja obtidos a respeito do assunto. A discussao tem como foco trabalhos que tratam da individualidade diretamente, assim como a analise de diferencas individuais em performances e a identificacao automatica de musicos. Primeiramente apresentamos uma reflexao a respeito do que e exatamente a individualidade na performance, e atraves de que meios ela pode ser comunicada/percebida. Em seguida sao discutidos alguns aspectos e desafios do estudo da individualidade na musica popular e erudita, dificuldades metodologicas e as diferencas tipicamente existentes en...
In order to map the spectral characteristics of the great variety of sounds a single musical inst... more In order to map the spectral characteristics of the great variety of sounds a single musical instrument may produce, different notes were performed and sampled in several intensity levels across the whole extension of a clarinet. Amplitude and frequency timevarying curves of partials were measured by Discrete Fourier Transform. A limited set of orthogonal spectral bases was derived by Principal Component Analysis techniques. These bases defined spectral subspaces capable of representing all tested sounds and of grouping of them, which were validated by Mean Opinion Score auditory tests of similarity. Subspaces involving larger groups of notes were used to compare the sounds according to the distance metrics of the representation.
This work presents a brief description of the human auditory system together with the history of ... more This work presents a brief description of the human auditory system together with the history of human comprehension of the auditory function, its main features, and classic models used to represent it. First, a historical view of the hearing apparatus is presented. After that, the physiology of the peripheral auditory system is described. The process of acoustic propagation through the outer, middle and inner ear, as well as the mechanism of transformation of cochlea inner hair cell motion into neuron spikes are explained. Next, Flanagan’s mathematical representation (based on physiological data acquired by von Bekesy) of the passive relation between the sound that reaches the outer ear and the motion of the cochlea basilar membrane. Flanagan’s model is followed by Lyon’s model of the cochlea, Meddis’ model of the inner hair cell, and Patterson’s Auditory Image Model. Finally, the IPEM Toolbox is introduced as an example of music analysis system that incorporates an auditory model ...
Uploads
Papers by Mauricio Alves Loureiro