Audio-visual deep learning methods for musical instrument classification and separation

In music perception, the information we receive from a visual system and audio system is often complementary. Moreover, visual perception plays an important role in the overall experience of being exposed to a music performance. This fact brings attention to machine learning methods that could combi...

ver descrição completa

Detalhes bibliográficos
Autor: Slizovskaia, Olga
Tipo de documento: tese
Estado:Versão publicada
Data de publicação:2020
País:España
Recursos:CBUC, CESCA
Repositório:TDR. Tesis Doctorales en Red
OAI Identifier:oai:www.tdx.cat:10803/669963
Acesso em linha:http://hdl.handle.net/10803/669963
Access Level:Acceso aberto
Palavra-chave:Audio-visual deep learning
Multimodal deep learning
Music information retrieval
Musical performance video
Musical performance analysis
Musical instrument classification
Sound source separation
Fusion techniques
Conditioning techniques
Aprendizaje profundo audiovisual
Aprendizaje profundo multimodal
Recuperación de información musical
Video musical
Análisis de interpretación musical
Clasificación de instrumentos musicales
Separación de fuentes de sonido
Técnicas de fusión
Técnicas de acondicionamiento
62
Descrição
Resumo:In music perception, the information we receive from a visual system and audio system is often complementary. Moreover, visual perception plays an important role in the overall experience of being exposed to a music performance. This fact brings attention to machine learning methods that could combine audio and visual information for automatic music analysis. This thesis addresses two research problems: instrument classification and source separation in the context of music performance videos. A multimodal approach for each task is developed using deep learning techniques to train an encoded representation for each modality. For source separation, we also study two approaches conditioned on instrument labels and examine the influence that two extra sources of information have on separation performance compared with a conventional model. Another important aspect of this work is in the exploration of different fusion methods which allow for better multimodal integration of information sources from associated domains.