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

Symbolic music is widely used in various deep learning tasks, includinggeneration, transcription, synthesis, and Music Information Retrieval (MIR). Itis mostly employed with discrete models like Transformers, which require musicto be tokenized, i.e., formatted into sequences of distinct elements calledtokens. Tokenization can be performed in different ways. As Transformer canstruggle at reasoning, but capture more easily explicit information, it isimportant to study how the way the information is represented for such modelimpact their performances. In this work, we analyze the common tokenizationmethods and experiment with time and note duration representations. We comparethe performances of these two impactful criteria on several tasks, includingcomposer and emotion classification, music generation, and sequencerepresentation learning. We demonstrate that explicit information leads tobetter results depending on the task.

Impact of time and note duration tokenizations on deep learning symbolic music modeling