Real-Time monophonic and polyphonic audio classification from power spectra

作者：

Highlights：

• Normalized power spectra are discriminant features for audio classification.

• Efficient modeling both in monophonic and polyphonic recognition.

• Computation time accuracy tradeoff using hierarchical clustering of the models.

• Outperform deep neural networks and other learning algorithms.

摘要

•Normalized power spectra are discriminant features for audio classification.•Efficient modeling both in monophonic and polyphonic recognition.•Computation time accuracy tradeoff using hierarchical clustering of the models.•Outperform deep neural networks and other learning algorithms.

论文关键词：Real-time,Audio classification,Machine learning,Monophonic,Polyphonic,Generative model,Nonparametric estimation

论文评审过程：Received 10 July 2018, Revised 8 March 2019, Accepted 21 March 2019, Available online 21 March 2019, Version of Record 26 March 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.03.017