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

Autor:	Fradet, Nathan, Gutowski, Nicolas, Chhel, Fabien, Briot, Jean-Pierre
Rok vydání:	2023
Předmět:	Computer Science - Sound Computer Science - Artificial Intelligence Computer Science - Machine Learning Electrical Engineering and Systems Science - Audio and Speech Processing
Druh dokumentu:	Working Paper
Popis:	Symbolic music is widely used in various deep learning tasks, including generation, transcription, synthesis, and Music Information Retrieval (MIR). It is mostly employed with discrete models like Transformers, which require music to be tokenized, i.e., formatted into sequences of distinct elements called tokens. Tokenization can be performed in different ways. As Transformer can struggle at reasoning, but capture more easily explicit information, it is important to study how the way the information is represented for such model impact their performances. In this work, we analyze the common tokenization methods and experiment with time and note duration representations. We compare the performances of these two impactful criteria on several tasks, including composer and emotion classification, music generation, and sequence representation learning. We demonstrate that explicit information leads to better results depending on the task. Comment: ISMIR 2023
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2310.08497 Zobrazit plný text záznamu View this record from Arxiv