Investigating CNN-based Instrument Family Recognition for Western Classical Music Recordings
Investigating CNN-based Instrument Family Recognition for Western Classical Music Recordings
复制标题
研究基于 CNN 的西方古典音乐录音乐器族识别
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Meinard Müller
中科院分区：
文献类型：
--
作者：
Michael Taenzer;J. Abeßer;S. I. Mimilakis;Christof Weiss;Meinard Müller
西方古典音乐包括为不同的合奏组成的丰富的曲目。通常，这些合奏团包括一个或两个家庭的乐器木管，铜管，钢琴，声乐，弦乐。在本文中，我们考虑的任务，自动识别乐器家庭的音乐录音。作为一个主要的贡献，我们调查的影响，数据规范化，预处理和增强技术的泛化能力的模型。我们报告的实验使用三个数据集的单音色录音涵盖不同层次的音色复杂性：孤立的音符，孤立的旋律，和复调作品。虽然数据增强和光谱补丁的归一化被证明是贝内的，但对数压缩和通道能量归一化等预处理策略并没有带来实质性的改善。此外，我们的跨数据集实验表明进一步优化例程（如域自适应）的必要性。
Western classical music comprises a rich repertoire composed for different ensembles. Often, these ensembles consist of instruments from one or two of the families wood-winds, brass, piano, vocals, and strings. In this paper, we consider the task of automatically recognizing instrument families from music recordings. As one main contribution, we investigate the influence of data normalization, pre-processing, and augmentation techniques on the generalization capability of the models. We report on experiments using three datasets of monotimbral recordings covering different levels of timbral complexity: isolated notes, isolated melodies, and polyphonic pieces. While data augmentation and the normalization of spectral patches turned out to be beneficial, pre-processing strategies such as logarithmic compression and channel-energy normalization did not lead to substantial improvements. Furthermore, our cross-dataset experiments indicate the necessity of further optimization routines such as domain adaptation.
每通道能量标准化：原因和方法
影响因子：
3.9
作者：
Lostanlen, Vincent;Salamon, Justin;Cartwright, Mark;McFee, Brian;Farnsworth, Andrew;Kelling, Steve;Bello, Juan Pablo
通讯作者：
Bello, Juan Pablo
MedleyDB：用于注释密集型 MIR 研究的多轨数据集
DOI：
--
发表时间：
2014
期刊：
15th International Society for Music Information Retrieval Conference
影响因子：
--
作者：
Bittner, R.
通讯作者：
Bittner, R.