Since the advent of real-time computer music environments, composers have increasingly incorporated DSP analysis, synthesis, and processing algorithms in their creative practices. Those processes became part of interactive systems that use real-time computational tools in musical compositions that explore diverse techniques to generate, spatialize, and process instrumental/vocal sounds. Parallel to the development of these tools and the expansion of DSP methods, new techniques focused on sound/musical information extraction became part of the tools available for music composition. In this context, this article discusses the creative use of Machine Listening and Musical Information Retrieval techniques applied in the composition of live-electronics works. By pointing out some practical applications and creative approaches, we aim to circumscribe, in a general way, the strategies for employing Machine Listening and Music Information Retrieval techniques observed in a set of live-electronics pieces, categorizing four compositional approaches: namely, mapping, triggering, scoring, and procedural paradigms of application of machine listening techniques in the context of live-electronics music compositions.
{"title":"Mapping, Triggering, Scoring, and Procedural Paradigms of Machine Listening Application in Live-Electronics Compositions","authors":"Vinicius César De Oliveira, J. Padovani","doi":"10.5753/sbcm.2021.19445","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19445","url":null,"abstract":"Since the advent of real-time computer music environments, composers have increasingly incorporated DSP analysis, synthesis, and processing algorithms in their creative practices. Those processes became part of interactive systems that use real-time computational tools in musical compositions that explore diverse techniques to generate, spatialize, and process instrumental/vocal sounds. Parallel to the development of these tools and the expansion of DSP methods, new techniques focused on sound/musical information extraction became part of the tools available for music composition. In this context, this article discusses the creative use of Machine Listening and Musical Information Retrieval techniques applied in the composition of live-electronics works. By pointing out some practical applications and creative approaches, we aim to circumscribe, in a general way, the strategies for employing Machine Listening and Music Information Retrieval techniques observed in a set of live-electronics pieces, categorizing four compositional approaches: namely, mapping, triggering, scoring, and procedural paradigms of application of machine listening techniques in the context of live-electronics music compositions.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125273822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, we introduce an original approach to computerized music analysis within the graphical computer-assisted composition environment called PWGL. Our aim is to facilitate the realtime analysis of interactive scores written in common Western music notation. To this end, we have developed a novel library that allows us to analyze scores realized with the help of ENP (the graphical music notation module of PWGL), and to visualize the results of the analysis in realtime. ENP is extended to support the display of supplementary information that can be drawn on top of the score as an overlay. The analysis backend is realized with the help of our builtin musical scripting language based on pattern matching. The analysis results are presented directly as a part of the original score leveraging the extensible and interactive visualization capabilities of ENP. In this paper we describe the current state of the library and present, as a case study, a fully functional application allowing for the realtime analysis and display of voice leading errors according to the counterpoint rules developed mainly in the Renaissance and Baroque eras.
{"title":"Towards Real-time Score Analysis in PWGL","authors":"Mika Kuuskankare","doi":"10.5753/sbcm.2021.19424","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19424","url":null,"abstract":"In this paper, we introduce an original approach to computerized music analysis within the graphical computer-assisted composition environment called PWGL. Our aim is to facilitate the realtime analysis of interactive scores written in common Western music notation. To this end, we have developed a novel library that allows us to analyze scores realized with the help of ENP (the graphical music notation module of PWGL), and to visualize the results of the analysis in realtime. ENP is extended to support the display of supplementary information that can be drawn on top of the score as an overlay. The analysis backend is realized with the help of our builtin musical scripting language based on pattern matching. The analysis results are presented directly as a part of the original score leveraging the extensible and interactive visualization capabilities of ENP. In this paper we describe the current state of the library and present, as a case study, a fully functional application allowing for the realtime analysis and display of voice leading errors according to the counterpoint rules developed mainly in the Renaissance and Baroque eras.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125517091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Regis Faria, Ricardo Thomasi, João Monnazzi, Eduardo Bonachela, André Giolito, Gabriel Lemos
This paper presents a concise report on the research developed at the Laboratory of Audio and Music Technology at the EACH-USP. The laboratory was founded in 2011 targeting the areas of music technology, musical acoustics and bioacoustics, strengthening its scope in 2019 to the areas of sound and music computing and audio engineering. Six projects are presented herein, describing their application areas, goals, achievements and perspectives.
{"title":"A prospective report on the research developed at the Laboratory of Audio and Music Technology at USP","authors":"Regis Faria, Ricardo Thomasi, João Monnazzi, Eduardo Bonachela, André Giolito, Gabriel Lemos","doi":"10.5753/sbcm.2021.19461","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19461","url":null,"abstract":"This paper presents a concise report on the research developed at the Laboratory of Audio and Music Technology at the EACH-USP. The laboratory was founded in 2011 targeting the areas of music technology, musical acoustics and bioacoustics, strengthening its scope in 2019 to the areas of sound and music computing and audio engineering. Six projects are presented herein, describing their application areas, goals, achievements and perspectives.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131974637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Micael Antunes, Guilherme Feulo do Espírito Santo, J. Manzolli
We present in this paper an implementation of a roughness descriptor based on the model of Pantelis Vassilakis. We apply the implemented descriptor to perform a computer-aided musical analysis of György Ligeti’s Continuum for harpsichord (1968). Our analysis establishes a parallel between the roughness model, concepts of sound mass composition, and Ligeti’s ideas of timbre of movement and permeability. As a result, we display a graphical representation for the roughness present in the Continuum’s recording, propose a formal segmentation for the piece, and analyze its formal development from a musicological point of view.
{"title":"A Roughness Model Implementation to Analyze Sound Mass Composition: A Case Study in Ligeti’s Continuum for Harpsichord (1968)","authors":"Micael Antunes, Guilherme Feulo do Espírito Santo, J. Manzolli","doi":"10.5753/sbcm.2021.19430","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19430","url":null,"abstract":"We present in this paper an implementation of a roughness descriptor based on the model of Pantelis Vassilakis. We apply the implemented descriptor to perform a computer-aided musical analysis of György Ligeti’s Continuum for harpsichord (1968). Our analysis establishes a parallel between the roughness model, concepts of sound mass composition, and Ligeti’s ideas of timbre of movement and permeability. As a result, we display a graphical representation for the roughness present in the Continuum’s recording, propose a formal segmentation for the piece, and analyze its formal development from a musicological point of view.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129487844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
F. Schiavoni, Ana Clara Montenegro Fonseca, Avner Maximiliano de Paulo, Carlos Edmundo Oliveira Souza, Cleisson Dener da Silva, Emerson Junio Silva Costa, Fábio dos Passos Carvalho, Gabriel Lopes Rocha, Gabriel Rodrigues Chaves Carneiro, Igino de Oliveira Silva Júnior, Igor Alves Estevam Andrade, Isadora Franco Oliveira, João Marcos Rodrigues Andrade Lara, João Teixeira Araújo, João Pedro Nunes Pereira de Oliveira, Jônatas Araújo da Silva, Luan Luiz Gonçalves, L. E. Santos, Luiz Gustavo Balzanelli Sousa, Mariana Pereira Lellis, Matheus de Bonfim Rodrigues Jordão, R. Andrade, Rebeca Lima Soares, Rômulo Augusto Vieira Costa, Samuel Rodrigues Rabay
O ALICE - Arts Lab in Interfaces, Computers, and Everything Else é um laboratório de pesquisa do Departamento de Computação da Federal University of São João del-Rei. Acolhendo pesquisadores de várias áreas, o ALICE tem desenvolvido pesquisas nas áreas de processos criativos amparados pela tecnologia, música em rede, paisagens sonoras, orquestras de dispositivos e outras áreas em que é possível aliar arte e tecnologia sem que uma destas áreas se coloque acima da outra. Neste artigo, apresentamos nossos desenvolvimentos recentes e lamentos atuais em um ano em que a distância de nossos corpos tenta ser superada pela onipresença das redes de computadores.
ALICE -界面、计算机和其他一切艺术实验室是sao joao del rei联邦大学计算机系的一个研究实验室。ALICE欢迎来自不同领域的研究人员，在技术支持的创作过程、网络音乐、声音景观、设备管弦乐队和其他领域开展了研究，在这些领域中，艺术和技术可以结合起来，而不需要一个领域凌驾于另一个领域之上。在这篇文章中，我们展示了我们最近的发展和当前的哀叹，在这一年里，我们的身体距离试图被无处不在的计算机网络所克服。
{"title":"Alice no país da pandemia (2021)","authors":"F. Schiavoni, Ana Clara Montenegro Fonseca, Avner Maximiliano de Paulo, Carlos Edmundo Oliveira Souza, Cleisson Dener da Silva, Emerson Junio Silva Costa, Fábio dos Passos Carvalho, Gabriel Lopes Rocha, Gabriel Rodrigues Chaves Carneiro, Igino de Oliveira Silva Júnior, Igor Alves Estevam Andrade, Isadora Franco Oliveira, João Marcos Rodrigues Andrade Lara, João Teixeira Araújo, João Pedro Nunes Pereira de Oliveira, Jônatas Araújo da Silva, Luan Luiz Gonçalves, L. E. Santos, Luiz Gustavo Balzanelli Sousa, Mariana Pereira Lellis, Matheus de Bonfim Rodrigues Jordão, R. Andrade, Rebeca Lima Soares, Rômulo Augusto Vieira Costa, Samuel Rodrigues Rabay","doi":"10.5753/sbcm.2021.19462","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19462","url":null,"abstract":"O ALICE - Arts Lab in Interfaces, Computers, and Everything Else é um laboratório de pesquisa do Departamento de Computação da Federal University of São João del-Rei. Acolhendo pesquisadores de várias áreas, o ALICE tem desenvolvido pesquisas nas áreas de processos criativos amparados pela tecnologia, música em rede, paisagens sonoras, orquestras de dispositivos e outras áreas em que é possível aliar arte e tecnologia sem que uma destas áreas se coloque acima da outra. Neste artigo, apresentamos nossos desenvolvimentos recentes e lamentos atuais em um ano em que a distância de nossos corpos tenta ser superada pela onipresença das redes de computadores.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128750913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Diego Furtado Silva, Micael Valterlânio da Silva, Ricardo Szram Filho, A. Silva
Music classification is one of the most studied tasks in music information retrieval. Notably, one of the targets with high interest in this task is the music genre. In this scenario, the use of deep neural networks has led to the current state-of-the-art results. Research endeavors in this knowledge domain focus on a single feature to represent the audio in the input for the classification model. Due to this task’s nature, researchers usually rely on time-frequency-based features, especially those designed to make timbre more explicit. However, the audio processing literature presents many strategies to build representations that reveal diverse characteristics of music, such as key and tempo, which may contribute with relevant information for the classification of genres. We showed an exploratory study on different neural network model fusion techniques for music genre classification with multiple features as input. Our results demonstrate that Multi-Feature Fusion Networks consistently improve the classification accuracy for suitable choices of input representations.
{"title":"On the Fusion of Multiple Audio Representations for Music Genre Classification","authors":"Diego Furtado Silva, Micael Valterlânio da Silva, Ricardo Szram Filho, A. Silva","doi":"10.5753/sbcm.2021.19423","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19423","url":null,"abstract":"Music classification is one of the most studied tasks in music information retrieval. Notably, one of the targets with high interest in this task is the music genre. In this scenario, the use of deep neural networks has led to the current state-of-the-art results. Research endeavors in this knowledge domain focus on a single feature to represent the audio in the input for the classification model. Due to this task’s nature, researchers usually rely on time-frequency-based features, especially those designed to make timbre more explicit. However, the audio processing literature presents many strategies to build representations that reveal diverse characteristics of music, such as key and tempo, which may contribute with relevant information for the classification of genres. We showed an exploratory study on different neural network model fusion techniques for music genre classification with multiple features as input. Our results demonstrate that Multi-Feature Fusion Networks consistently improve the classification accuracy for suitable choices of input representations.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124630658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pierre Schaeffers typomorphology (1966) proposes seven criteria of musical perception for the qualification of sound objects, which form the basis of his musical theory. This Solfège fits well into contexts where pitch is not the dominant dimension. Relying on similarities between the practice of reduced listening and the utilization of low-level audio descriptors, we present the first version of a real-time setup in which these descriptors are applied to qualify percussive sounds. The paper describes the tools and strategies used for addressing different criteria: envelope followers with different window sizes and filtering; detection of transients and amplitude modulations; extraction and counting of spectral components; estimation of intrinsic dissonance and spectral distribution; among others. The extracted data is subjected to simple statistical analysis, producing scalar values associated with each segmented object. Finally, we present a variety of examples.
Pierre Schaeffers的类型学(typomorphology, 1966)为声音对象的资格提出了七个音乐感知标准，这些标准构成了他的音乐理论的基础。在音高不是主要维度的情况下，这种唱法很适合。依靠减少听力的实践和低级音频描述符的使用之间的相似性，我们提出了一个实时设置的第一个版本，其中这些描述符被应用于限定打击声音。本文描述了用于解决不同标准的工具和策略:具有不同窗口大小和过滤的信封追随者;瞬态和调幅检测;光谱成分的提取与计数;内禀不谐估计及谱分布;等等。提取的数据进行简单的统计分析，生成与每个分割对象相关联的标量值。最后，我们给出了各种各样的例子。
{"title":"Real-time Qualification of Percussive Sounds Based on Correspondences Between Schaeffer’s Solfège and Low-level Audio Descriptors","authors":"Sérgio Freire, J. Padovani, C. Campos","doi":"10.5753/sbcm.2021.19429","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19429","url":null,"abstract":"Pierre Schaeffers typomorphology (1966) proposes seven criteria of musical perception for the qualification of sound objects, which form the basis of his musical theory. This Solfège fits well into contexts where pitch is not the dominant dimension. Relying on similarities between the practice of reduced listening and the utilization of low-level audio descriptors, we present the first version of a real-time setup in which these descriptors are applied to qualify percussive sounds. The paper describes the tools and strategies used for addressing different criteria: envelope followers with different window sizes and filtering; detection of transients and amplitude modulations; extraction and counting of spectral components; estimation of intrinsic dissonance and spectral distribution; among others. The extracted data is subjected to simple statistical analysis, producing scalar values associated with each segmented object. Finally, we present a variety of examples.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125553122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Douglas Silva, Lucas Zampar, Felipe Rodrigues, C. Gomes
Este artigo apresenta ações em andamento e finalizadas pelo grupo itinerante de COmputacão mUsical e Tecnologias Emergentes (COUT-E) da Universidade Federal do Amapá (UNIFAP) durante o período de 2020 e 2021. Comenta-se sobre os projetos Handy Hear, CongAr, Acalma-te, classificação de ritmos populares da amazônia e sincronização rítmica entre fontes distintas em tempo real.
{"title":"COUT-e e práticas itinerantes durante 2020 e 2021","authors":"Douglas Silva, Lucas Zampar, Felipe Rodrigues, C. Gomes","doi":"10.5753/sbcm.2021.19464","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19464","url":null,"abstract":"Este artigo apresenta ações em andamento e finalizadas pelo grupo itinerante de COmputacão mUsical e Tecnologias Emergentes (COUT-E) da Universidade Federal do Amapá (UNIFAP) durante o período de 2020 e 2021. Comenta-se sobre os projetos Handy Hear, CongAr, Acalma-te, classificação de ritmos populares da amazônia e sincronização rítmica entre fontes distintas em tempo real.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124569334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A globalização afeta a preferência musical da sociedade atual, que apresenta continua estima por ou gêneros musicais internacionais em detrimento aos nacionais ou locais. A música é dos meios de comunicação utilizados para a construção e organização da estrutura social influenciando o estilo de vida, gostos e convivência interpessoal. Portanto, cada manifestação de gênero musical apresenta função distinta para um ouvinte, como por exemplo, para dançar, festejar, descançar, ajudar na solidão, na tristeza etc. Diversos aplicativos como spotify e soundcloud, utilizam-se de classificadores de gêneros musicais para indicar, prever ou sugerir novas músicas para seus ouvintes. Por diversos motivos, a maioria dos classificadores não possui informações de gêneros musicais regionais. Este trabalho propõem um modelo de classificação automático de gêneros musicais populares amazônicos. Inicialmente, confeccionou-se uma base de dados contendo os gêneros musicais: andino, brega, carimbó, cúmbia, marabaixo, pasillo, salsa e vaqueirada, oriundos da região da Amazônia Legal dos países: Brasil, Guiana Francesa, Venezuela, Colômbia, Equador, Bolívia e Peru. Para a construção da base de dados, extraiu-se diversas características de cada música ao total de 64 parâmetros. Analisou-se os modelos de aprendizado de máquina na qual XGB, KNN, SVM e MLP obtiveram acurácia de 67.62%, 74.12%, 71.35%, 76.13%, respectivamente.
{"title":"Modelo automático de classificação de gêneros musicais amazônicos","authors":"D. Silva, Lucas Zampar, F. Rodrigues, C. Gomes","doi":"10.5753/sbcm.2021.19453","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19453","url":null,"abstract":"A globalização afeta a preferência musical da sociedade atual, que apresenta continua estima por ou gêneros musicais internacionais em detrimento aos nacionais ou locais. A música é dos meios de comunicação utilizados para a construção e organização da estrutura social influenciando o estilo de vida, gostos e convivência interpessoal. Portanto, cada manifestação de gênero musical apresenta função distinta para um ouvinte, como por exemplo, para dançar, festejar, descançar, ajudar na solidão, na tristeza etc. Diversos aplicativos como spotify e soundcloud, utilizam-se de classificadores de gêneros musicais para indicar, prever ou sugerir novas músicas para seus ouvintes. Por diversos motivos, a maioria dos classificadores não possui informações de gêneros musicais regionais. Este trabalho propõem um modelo de classificação automático de gêneros musicais populares amazônicos. Inicialmente, confeccionou-se uma base de dados contendo os gêneros musicais: andino, brega, carimbó, cúmbia, marabaixo, pasillo, salsa e vaqueirada, oriundos da região da Amazônia Legal dos países: Brasil, Guiana Francesa, Venezuela, Colômbia, Equador, Bolívia e Peru. Para a construção da base de dados, extraiu-se diversas características de cada música ao total de 64 parâmetros. Analisou-se os modelos de aprendizado de máquina na qual XGB, KNN, SVM e MLP obtiveram acurácia de 67.62%, 74.12%, 71.35%, 76.13%, respectivamente.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131864674","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
O processamento de sinais de áudio em conjunto com modelos de aprendizado de máquina tem aplicações em diversas áreas: música, análise forense e análise de fala humana e ruído ambiente. Analisar música de diferentes gêneros pode encorajar investigações interessantes pela comunidade científica, como a investigação de tendências culturais. Este trabalho apresenta uma iniciativa para o processamento de canções vocaloides que ganharam grande popularidade nas redes sociais. Os classificadores Support Vector Machine (SVM) foram treinados em dois experimentos para distinguir canções da vocaloide Hatsune Miku de canções instrumentais e canções de outros vocaloides, apresentando resultados promissores de precisão acima de 80%, que validam a iniciativa.
{"title":"Identificação de Áudio Vocaloid com Support Vector Machines: um Estudo de Caso da Hatsune Miku","authors":"F. Almeida, V. Hayashi","doi":"10.5753/sbcm.2021.19451","DOIUrl":"https://doi.org/10.5753/sbcm.2021.19451","url":null,"abstract":"O processamento de sinais de áudio em conjunto com modelos de aprendizado de máquina tem aplicações em diversas áreas: música, análise forense e análise de fala humana e ruído ambiente. Analisar música de diferentes gêneros pode encorajar investigações interessantes pela comunidade científica, como a investigação de tendências culturais. Este trabalho apresenta uma iniciativa para o processamento de canções vocaloides que ganharam grande popularidade nas redes sociais. Os classificadores Support Vector Machine (SVM) foram treinados em dois experimentos para distinguir canções da vocaloide Hatsune Miku de canções instrumentais e canções de outros vocaloides, apresentando resultados promissores de precisão acima de 80%, que validam a iniciativa.","PeriodicalId":292360,"journal":{"name":"Anais do XVIII Simpósio Brasileiro de Computação Musical (SBCM 2021)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115373110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：