The DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Multimodal Metric Learning for Tag-Based Music Retrieval
Open accessAbstract
Tag-based music retrieval is crucial to browse large-scale mu-sic libraries efficiently. Hence, automatic music tagging has been actively explored, mostly as a classification task, which has an inherent limitation: a fixed vocabulary. On the other hand, metric learning enables flexible vocabularies by using pretrained word embeddings as side information. Also, met-ric learning has proven its suitability for cross-modal retrieval tasks in other domains (e.g., text-to-image) by jointly learning a multimodal embedding space. In this paper, we investigate three ideas to successfully introduce multimodal metric learning for tag-based music retrieval: elaborate triplet sampling, acoustic and cultural music information, and domain-specific word embeddings. Our experimental results show that the proposed ideas enhance the retrieval system quantitatively and qualitatively. Furthermore, we release the MSD500: a subset of the Million Song Dataset (MSD) containing 500 cleaned tags, 7 manually annotated tag categories, and user taste profiles.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Bertin-Mahieux, T., Ellis, D. P. W., Whitman, B., & Lamere, P. (2011) · Columbia University · doi:10.7916/d8nz8j07 · Cited by 944
Flagged first, then by citations
Research works printed but not linked: 20
- hoffer, 2015, Deep metric learning using triplet network, International Workshop on Similarity-Based Pattern Recognition, 10.1007/978-3-319-24261-3_7 · DOI · Paper page
- 10.3115/v1/D14-1162 · DOI · Paper page
- mikolov, 2013, Efficient estimation of word representations in vector space, International Conference on Learning Representations
- pachet, 2005, Knowledge management and musical metadata, Idea Group, 12
- 10.1145/358916.358995 · DOI · Paper page
- 10.1109/ICDM.2008.22 · DOI
- 10.1145/2187980.2188222 · DOI
- korzeniowski, 2018, Mood classification using listening data, Proc of the 19th International Society for Music Information Retrieval Conference
- koch, 2015, Siamese neural networks for one-shot image recognition, ICML Deep Learning Workshop
- weinberger, 2009, Distance metric learning for large margin nearest neighbor classification, Journal of Machine Learning Research, 10
- 10.5334/tismir.10 · DOI
- 10.1109/CVPR.2016.541 · DOI
- surís, 2018, Cross-modal embed-dings for video and audio retrieval, Proc of the European Conference on Computer Vision
- frome, 2013, Devise: A deep visual-semantic embedding model, Advances in neural information processing systems
- xing, 2003, Distance metric learning with application to clustering with side-information, Advances in neural information processing systems
- won, 2020, Evaluation of cnn-based automatic music tagging models, Proc of the 17th Sound and Music Computing
- choi, 2019, Zero-shot learning for audio-based music classification and tagging, Proc of the 20th International Society for Music Information Retrieval Conference
- wu, 2017, Sampling matters in deep embed-ding learning, Proc of the IEEE International Conference on Computer Vision
- 10.1109/CVPR.2015.7298682 · DOI · Paper page
- 10.1109/ICCV.2015.22 · DOI
Cited by
Sorted by citations
Enriched music representations with multiple cross-modal contrastive learning
Ferraro, A., Favory, X., Drossos, K., Kim, Y., & Bogdanov, D. (2021) · IEEE Signal Processing Letters · doi:10.1109/lsp.2021.3071082 · Cited by 23
LP-MusicCaps: LLM-Based Pseudo Music Captioning
Doh, S., Choi, K., Lee, J., & Nam, J. (2023) · arXiv · doi:10.48550/arxiv.2307.16372 · Cited by 16
“We adopt a balanced sampling [38], which uniformly samples an anchor tag first and then selects an annotated item.”
Method
Metadata
- Authors
- Won, M., Oramas, S., Nieto, O., Gouyon, F., & Serra, X.
- Affiliations
- Universitat Pompeu Fabra (ES)
- Pandora (United States) (US)
- Venue
- ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2021-06-06
- Pages
- 591-595
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W3162583214
- Cited by
- 33
- References
- 2
- Citations by year
- 2025: 4 · 2024: 9 · 2023: 8 · 2022: 4 · 2021: 8
- FWCI
- 4.41
- Type
- conference paper
- Language
- en
- Open access
- green
- hdl.handle.net
- Licence
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion)
- Repositori digital de la UPF (Universitat Pompeu Fabra) (submittedVersion)
- Funders
- Ministry of Economy
- Topics
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Physical Sciences › Computer Science › Information Systems › Recommender Systems and Techniques
- Social Sciences › Arts and Humanities › Music › Music Technology and Sound Studies
- Keywords
- metric learning, multimodal embeddings, word embeddings, acoustic features, Million Song Dataset
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
