English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present Mirable's submission to the 2021 Emotions and Themes in Musicchallenge. In this work, we intend to address the question: can we leveragesemi-supervised learning techniques on music emotion recognition? With that, weexperiment with noisy student training, which has improved model performance inthe image classification domain. As the noisy student method requires a strongteacher model, we further delve into the factors including (i) input traininglength and (ii) complementary music representations to further boost theperformance of the teacher model. For (i), we find that models trained withshort input length perform better in PR-AUC, whereas those trained with longinput length perform better in ROC-AUC. For (ii), we find that using harmonicpitch class profiles (HPCP) consistently improve tagging performance, whichsuggests that harmonic representation is useful for music emotion tagging.Finally, we find that noisy student method only improves tagging results forthe case of long training length. Additionally, we find that ensemblingrepresentations trained with different training lengths can improve taggingresults significantly, which suggest a possible direction to exploreincorporating multiple temporal resolutions in the network architecture forfuture work.

Semi-supervised music emotion recognition using noisy student training and harmonic pitch class profiles