Novel Audio Features for Music Emotion Recognition
Main Author: | |
---|---|
Publication Date: | 2020 |
Other Authors: | , |
Format: | Article |
Language: | eng |
Source: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Download full: | http://hdl.handle.net/10316/94071 https://doi.org/10.1109/TAFFC.2018.2820691 |
Summary: | This work advances the music emotion recognition state-of-the-art by proposing novel emotionally-relevant audio features. We reviewed the existing audio features implemented in well-known frameworks and their relationships with the eight commonly defined musical concepts. This knowledge helped uncover musical concepts lacking computational extractors, to which we propose algorithms - namely related with musical texture and expressive techniques. To evaluate our work, we created a public dataset of 900 audio clips, with subjective annotations following Russell's emotion quadrants. The existent audio features (baseline) and the proposed features (novel) were tested using 20 repetitions of 10-fold cross-validation. Adding the proposed features improved the F1-score to 76.4 percent (by 9 percent), when compared to a similar number of baseline-only features. Moreover, analysing the features relevance and results uncovered interesting relations, namely the weight of specific features and musical concepts to each emotion quadrant, and warrant promising new directions for future research in the field of music emotion recognition, interactive media, and novel music interfaces. |
id |
RCAP_c06fe4c3414ea61b4b14a5200a955f95 |
---|---|
oai_identifier_str |
oai:estudogeral.uc.pt:10316/94071 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Novel Audio Features for Music Emotion RecognitionAffective computingAudio databasesEmotion recognitionFeature extractionMusic information retrievalThis work advances the music emotion recognition state-of-the-art by proposing novel emotionally-relevant audio features. We reviewed the existing audio features implemented in well-known frameworks and their relationships with the eight commonly defined musical concepts. This knowledge helped uncover musical concepts lacking computational extractors, to which we propose algorithms - namely related with musical texture and expressive techniques. To evaluate our work, we created a public dataset of 900 audio clips, with subjective annotations following Russell's emotion quadrants. The existent audio features (baseline) and the proposed features (novel) were tested using 20 repetitions of 10-fold cross-validation. Adding the proposed features improved the F1-score to 76.4 percent (by 9 percent), when compared to a similar number of baseline-only features. Moreover, analysing the features relevance and results uncovered interesting relations, namely the weight of specific features and musical concepts to each emotion quadrant, and warrant promising new directions for future research in the field of music emotion recognition, interactive media, and novel music interfaces.This work was supported by the MOODetector project (PTDC/EIA-EIA/102185/2008), financed by the Fundação para Ciência e a Tecnologia (FCT) and Programa Operacional Temático Factores de Competitividade (COMPETE) – Portugal, as well as the PhD Scholarship SFRH/BD/91523/ 2012, funded by the Fundação para Ciência e a Tecnologia (FCT), Programa Operacional Potencial Humano (POPH) and Fundo Social Europeu (FSE).IEEE2020info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articlehttp://hdl.handle.net/10316/94071http://hdl.handle.net/10316/94071https://doi.org/10.1109/TAFFC.2018.2820691eng1949-30452371-9850https://ieeexplore.ieee.org/document/8327886/Panda, RenatoMalheiro, RicardoPaiva, Rui Pedroinfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2021-05-25T07:35:57Zoai:estudogeral.uc.pt:10316/94071Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-19T21:12:53.739405Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Novel Audio Features for Music Emotion Recognition |
title |
Novel Audio Features for Music Emotion Recognition |
spellingShingle |
Novel Audio Features for Music Emotion Recognition Panda, Renato Affective computing Audio databases Emotion recognition Feature extraction Music information retrieval |
title_short |
Novel Audio Features for Music Emotion Recognition |
title_full |
Novel Audio Features for Music Emotion Recognition |
title_fullStr |
Novel Audio Features for Music Emotion Recognition |
title_full_unstemmed |
Novel Audio Features for Music Emotion Recognition |
title_sort |
Novel Audio Features for Music Emotion Recognition |
author |
Panda, Renato |
author_facet |
Panda, Renato Malheiro, Ricardo Paiva, Rui Pedro |
author_role |
author |
author2 |
Malheiro, Ricardo Paiva, Rui Pedro |
author2_role |
author author |
dc.contributor.author.fl_str_mv |
Panda, Renato Malheiro, Ricardo Paiva, Rui Pedro |
dc.subject.por.fl_str_mv |
Affective computing Audio databases Emotion recognition Feature extraction Music information retrieval |
topic |
Affective computing Audio databases Emotion recognition Feature extraction Music information retrieval |
description |
This work advances the music emotion recognition state-of-the-art by proposing novel emotionally-relevant audio features. We reviewed the existing audio features implemented in well-known frameworks and their relationships with the eight commonly defined musical concepts. This knowledge helped uncover musical concepts lacking computational extractors, to which we propose algorithms - namely related with musical texture and expressive techniques. To evaluate our work, we created a public dataset of 900 audio clips, with subjective annotations following Russell's emotion quadrants. The existent audio features (baseline) and the proposed features (novel) were tested using 20 repetitions of 10-fold cross-validation. Adding the proposed features improved the F1-score to 76.4 percent (by 9 percent), when compared to a similar number of baseline-only features. Moreover, analysing the features relevance and results uncovered interesting relations, namely the weight of specific features and musical concepts to each emotion quadrant, and warrant promising new directions for future research in the field of music emotion recognition, interactive media, and novel music interfaces. |
publishDate |
2020 |
dc.date.none.fl_str_mv |
2020 |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/10316/94071 http://hdl.handle.net/10316/94071 https://doi.org/10.1109/TAFFC.2018.2820691 |
url |
http://hdl.handle.net/10316/94071 https://doi.org/10.1109/TAFFC.2018.2820691 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
1949-3045 2371-9850 https://ieeexplore.ieee.org/document/8327886/ |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
IEEE |
publisher.none.fl_str_mv |
IEEE |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799134024989409280 |