http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
A Comparative Study of LSTM and Transformer Models in Music Melody Generation
郑嘉祥(Jiaxiang ZHENG),墨曦(Moxi CAO) Korea Institute for Humanities and Social Sciences 2023 Journal of Global Arts Studies (JGAS) Vol.1 No.1
【背景】近年来,利用深度学习模型生成音乐已经发展为AI 音乐的主流方向,但音乐生成任务的主流模型仍然存在着一些问题,其中最大的问题就是不能有效地模拟音乐结构,使计算机创作出符合音乐结构的乐曲。【目的】为此,我们需要探究哪些模型能够很好地模拟音乐的结构,创造更加人性化的音乐。【方法】我们通过对比实验,对比LSTM 与Transformer 模型所生成音乐的优点与缺点,并在此基础上提出改进方案。【结果】实验结果证明,LSTM 在较短的序列上模拟音乐结构的表现优于Transformer,但其无法处理过长的序列;而Transformer 在处理较长序列的表现优于LSTM,并通过改进后能在较长的序列上有效地模拟音乐结构,创作出符合人类音乐听觉的乐曲。【结论】因此我们认为,Transformer 模型更加适合AI 音乐作曲任务,并在未来通过改进其注意力机制来提高音乐结构的识别能力是音乐生成的主流方向。 [Background] In recent years, using deep learning models to generate music has become the mainstream direction in AI music. However, the main models for music generation still face several issues, the biggest of which is the inability to effectively simulate musical structure, hindering computers from creating compositions that conform to musical structures. [Objective] To address this, we need to explore which models can effectively simulate the structure of music and create more humanized music. [Method]We conduct comparative experiments, analyzing the advantages and disadvantages of music generated by LSTM and Transformer models, and propose improvements based on the findings. [Results] Experimental results demonstrate that LSTM performs better than Transformer in simulating musical structure in shorter sequences, but struggles with longer sequences; whereas Transformer outperforms LSTM in handling longer sequences and can effectively simulate musical structure in longer sequences after improvements, creating compositions that align with human musical perception. [Conclusion]Therefore, we believe that the Transformer model is more suitable for AI music composition tasks, and improving its attention mechanism to enhance recognition of musical structure will be the mainstream direction for music generation in the future.
Seohyun Park,Zheng Jiaxiang,Jihye Lee,Jae-Joon Lee,Jung-Heun Ha 한국식품영양과학회 2021 한국식품영양과학회 학술대회발표집 Vol.2021 No.10
PUFAs has been reported to alleviate diet-induced systemic inflammation, dyslipidemia in mammals. To understand the effects of replacement of saturated fatty acids (SFAs) with PUFAs to obese subjects, we supplemented a normal diet with perilla oil (PO) or corn oil (CO) to genetically obese mice for 12 weeks with LPS injection. 1) PUFAs improved lipid profiles. PUFAs consumption attenuated hepatic TG and LDL-C; however, HDL-C was significantly increased in PO or CO replaced than normal chow consumed group. 2) Intriguingly, PUFAs consumption decreased inflammatory responses in the epididymal adipose tissue(EAT). PUFAs decreased the secretion of adipokines (IL-1β, IL-6 and CXCL1). Attenuation of adipocyte inflammation maybe related to the post-translational modification by decreasing phosphorylation of MAPKs. Prolonged inflammation may trigger endoplasmic reticulum (ER) stress, therefore, we logically postulated that consumption of PUFAs may decrease ER stress. 3) As our expectation, PUFAs consumption attenuated LPS-inducible ER stress in the EAT by attenuation of BiP and CHOP. Therefore, PUFAs decreased hepatic lipids accumulation, inflammation and ER stress in the EAT.