رسالة جامعية
Music similarity models applied to cover song identification and classification ; Modelos de similaridade musical aplicados à identificação e classificação de covers
| العنوان: | Music similarity models applied to cover song identification and classification ; Modelos de similaridade musical aplicados à identificação e classificação de covers |
|---|---|
| المؤلفون: | Bodo, Roberto Piassi Passos |
| المساهمون: | Queiroz, Marcelo Gomes de |
| بيانات النشر: | Biblioteca Digital de Teses e Dissertações da USP Universidade de São Paulo Instituto de Matemática e Estatística |
| سنة النشر: | 2021 |
| المجموعة: | University of São Paulo: Digital Library of Theses and Dissertations / Biblioteca Digital de Teses e Dissertações da USP |
| مصطلحات موضوعية: | Classificação de covers, Recuperação de informação musical, Identificação de covers, Computação musical, Similaridade musical, Cover song classification, Cover song identification, Computer music, Music information retrieval, Music similarity |
| الوصف: | The spread of digital music allowed the appearance of datasets with millions of music files. The processing of this huge number of audio files is carried out with techniques of Music Information Retrieval (MIR) that work directly with the audio content. The MIR task of most interest in this project is the modelling of Music Similarity. Our proposed approach follows this pipeline: extract audio features, aggregate local features into global features, and compute the similarities of every pair of songs from the dataset being processed. According to this approach, a triple {extractor_i, aggregator_j, distance_k} defines a music similarity model, and our main goal is to investigate the ability of similarity models to distinguish audio files from different classes. The music similarity models are also used to address specific problems such as Cover Song Identification (CSI), which is an MIR application related to Music Similarity, and the closely-related Cover Song Classification (CSC) problem. MIR-related techniques, such as Dataset Modifications and Matrix Fusion, are explored in the context of improving the results of music similarity models. This work presents several contributions, among which a comprehensive benchmark of music similarity models; the definition of new similarity matrices within CSC as a solution approach to the CSI problem; the exploration of different types of dataset modifications and an investigation of their effect on music similarity metrics; and the fusion of similarity matrices computed from secondary datasets obtained via source separation. The experiments presented produced encouraging results, indicating that the methods proposed in this thesis point towards novel approaches to Music Similarity that are worth further investigation and development. ; A difusão da música digital permitiu o surgimento de conjuntos de dados com milhões de arquivos de áudio. O processamento deste enorme número de arquivos é realizado com técnicas de Recuperação de Informação Musical que analisam diretamente ... |
| نوع الوثيقة: | doctoral or postdoctoral thesis |
| وصف الملف: | application/pdf |
| اللغة: | English |
| العلاقة: | https://www.teses.usp.br/teses/disponiveis/45/45134/tde-27092021-104421Test/ |
| DOI: | 10.11606/T.45.2021.tde-27092021-104421 |
| الإتاحة: | https://doi.org/10.11606/T.45.2021.tde-27092021-104421Test https://www.teses.usp.br/teses/disponiveis/45/45134/tde-27092021-104421Test/ |
| حقوق: | Liberar o conteúdo para acesso público. |
| رقم الانضمام: | edsbas.5D984B8D |
| قاعدة البيانات: | BASE |
| DOI: | 10.11606/T.45.2021.tde-27092021-104421 |
|---|