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

Pre-trained language models have achieved impressive results in various musicunderstanding and generation tasks. However, existing pre-training methods forsymbolic melody generation struggle to capture multi-scale, multi-dimensionalstructural information in note sequences, due to the domain knowledgediscrepancy between text and music. Moreover, the lack of available large-scalesymbolic melody datasets limits the pre-training improvement. In this paper, wepropose MelodyGLM, a multi-task pre-training framework for generating melodieswith long-term structure. We design the melodic n-gram and long span samplingstrategies to create local and global blank infilling tasks for modeling thelocal and global structures in melodies. Specifically, we incorporate pitchn-grams, rhythm n-grams, and their combined n-grams into the melodic n-gramblank infilling tasks for modeling the multi-dimensional structures inmelodies. To this end, we have constructed a large-scale symbolic melodydataset, MelodyNet, containing more than 0.4 million melody pieces. MelodyNetis utilized for large-scale pre-training and domain-specific n-gram lexiconconstruction. Both subjective and objective evaluations demonstrate thatMelodyGLM surpasses the standard and previous pre-training methods. Inparticular, subjective evaluations show that, on the melody continuation task,MelodyGLM gains average improvements of 0.82, 0.87, 0.78, and 0.94 inconsistency, rhythmicity, structure, and overall quality, respectively.Notably, MelodyGLM nearly matches the quality of human-composed melodies on themelody inpainting task.

MelodyGLM: Multi-task Pre-training for Symbolic Melody Generation