Litcius/Paper detail

Bytecover2: Towards Dimensionality Reduction of Latent Embedding for Efficient Cover Song Identification

Xingjian Du, Ke Chen, Zijie Wang, Bilei Zhu, Zejun Ma

2022ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)22 citationsDOI

Abstract

Convolutional neural network (CNN)-based methods have dominated the recent research of cover song identification (CSI). A typical example is the ByteCover system we proposed, which has achieved state-of-the-art results on all the mainstream datasets of CSI. In this paper, we propose an up-graded version of ByteCover, termed ByteCover2, which further improves ByteCover in both identification performance and efficiency. Compared with ByteCover, ByteCover2 is designed with an additional PCA-FC module, which integrates the capability of principal component analysis (PCA) and fully-connected (FC) neural network for dimensionality reduction of the audio embedding, allowing ByteCover2 to perform CSI in a more precise and efficient way. We evaluated ByteCover2 on multiple datasets in different dimension sizes and training settings, where ByteCover2 beat all the compared methods including ByteCover, even with a dimension size of 128, which is 15 times smaller than that of ByteCover.

Topics & Concepts

Dimensionality reductionPrincipal component analysisComputer scienceEmbeddingConvolutional neural networkPattern recognition (psychology)Identification (biology)Cover (algebra)Artificial intelligenceDimension (graph theory)Artificial neural networkReduction (mathematics)Curse of dimensionalityMathematicsEngineeringGeometryBiologyBotanyPure mathematicsMechanical engineeringMusic and Audio ProcessingAnimal Vocal Communication and BehaviorSpeech and Audio Processing