Computer Analysis of Audio and Music

音频和音乐的计算机分析

• 批准号: 1000230741-2014
• 负责人: Tzanetakis, George
• 金额: $ 7.29万
• 依托单位: University of Victoria
• 依托单位国家: 加拿大
• 项目类别: Canada Research Chairs
• 财政年份: 2017
• 资助国家: 加拿大
• 起止时间: 2017-01-01 至 2018-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=642732
• 关键词: Computer; Analysis; Audio; Music

The central focus of this proposal is the development of algorithms and tools to analyze sound with specific emphasis on music. The first theme is the exploration of deep learning techniques, a family of learning algorithms that are currently showing a lot of promise, for dynamic, multi-task, semi-supervised learning in the context of music analysis. The second theme is the development of better frameworks and programming language abstractions for representing time and probabilities in the development of multi-modal interactive systems. The third theme explore the use of music information retrieval techniques in the area of computer-assisted music pedagogy. The findings of this research have the potential to have significant impact in the rapidly expanding use of computer music technology.

该提案的中心重点是开发算法和工具，以分析声音，特别强调音乐。第一个主题是深度学习技术的探索，这是一系列学习算法，目前在音乐分析的背景下显示出很大的前景，用于动态，多任务，半监督学习。第二个主题是开发更好的框架和编程语言抽象，用于在多模态交互系统的开发中表示时间和概率。第三个主题是探讨音乐资讯检索技术在电脑辅助音乐教学领域的应用。这项研究的结果有可能对迅速扩大的计算机音乐技术的使用产生重大影响。