Linear Transformer-GAN: A Novel Architecture to Symbolic Music Generation

鉴别器 计算机科学 变压器 生成语法 语音识别 人工智能 电压 电信 物理 量子力学 探测器
作者
Donglan Tian,Jinyan Chen,Xiangpeng Gao,Gang Pan
出处
期刊:Lecture Notes in Computer Science 卷期号:: 451-463
标识
DOI:10.1007/978-3-031-44195-0_37
摘要

Long-structured music generation that can be compared to human compositions remains an unresolved area of research. Since their introduction, the Transformer model and its variations, which rely on self-attention, have gained popularity in generating long-structured music. However, these models employ the teacher-forcing approach during training, which causes an exposure bias problem. Consequently, the generative model is incapable of producing music that consistently adheres to music theory. To address this issue, we propose a new Linear Transformer-GAN structure that generates high-quality music using a discriminator that has been trained to detect exposure bias. The Linear Transformer, a new and efficient variation of transformers, is creatively integrated with a generative adversarial network (GAN) to form our proposed model. In order to overcome the limitations of discrete domain data in GAN, we use the Policy Gradient and present a new discriminator structure that evaluates the current sequence reward based on several dimensions of music information. We use both the cross-entropy loss of different information dimensions and a music-theoretic mechanism to train the discriminator. Our experiments demonstrate that the proposed model generates music more consistent with music theory and is perceived as more pleasurable by listeners. This conclusion is supported by objective metrics and human evaluation. Overall, our approach offers a promising solution to the exposure bias problem in long-structured music generation and provides a more effective means of generating music that adheres to established music theory principles.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
Stephen发布了新的文献求助10
刚刚
科目三应助fly采纳,获得10
刚刚
隐形元绿完成签到,获得积分10
1秒前
脑洞疼应助蓝荆采纳,获得10
1秒前
赘婿应助Cc采纳,获得10
2秒前
2秒前
Akim应助滚筒洗衣机采纳,获得10
2秒前
3秒前
俏皮蜜蜂完成签到,获得积分10
3秒前
思源应助瓜子采纳,获得10
3秒前
LIUjun发布了新的文献求助10
3秒前
天天快乐应助关显锋采纳,获得10
3秒前
guozizi发布了新的文献求助10
4秒前
隐形元绿发布了新的文献求助10
4秒前
玩是罪恶的完成签到,获得积分10
4秒前
东郭井完成签到,获得积分10
5秒前
尧南完成签到,获得积分10
5秒前
彧辰完成签到 ,获得积分10
5秒前
6秒前
神奇宝贝完成签到,获得积分10
6秒前
6秒前
7秒前
Rainbow完成签到,获得积分10
7秒前
tutu完成签到,获得积分10
7秒前
ZY发布了新的文献求助10
7秒前
7秒前
最棒哒完成签到 ,获得积分10
8秒前
zyzy1996完成签到,获得积分20
8秒前
Hello应助狂野的蜡烛采纳,获得30
8秒前
罗亚亚完成签到,获得积分10
9秒前
9秒前
英俊的铭应助发飙的牛采纳,获得10
9秒前
脑洞疼应助宇宙甜心采纳,获得10
9秒前
9秒前
量子星尘发布了新的文献求助10
10秒前
10秒前
我是成坤发布了新的文献求助10
10秒前
GGGYQ完成签到 ,获得积分10
10秒前
zyzy1996发布了新的文献求助10
11秒前
11秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Cognitive Neuroscience: The Biology of the Mind 1000
Technical Brochure TB 814: LPIT applications in HV gas insulated switchgear 1000
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Picture Books with Same-sex Parented Families: Unintentional Censorship 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3969322
求助须知:如何正确求助?哪些是违规求助? 3514152
关于积分的说明 11172188
捐赠科研通 3249407
什么是DOI,文献DOI怎么找? 1794832
邀请新用户注册赠送积分活动 875437
科研通“疑难数据库(出版商)”最低求助积分说明 804781