MDCNN-SID: Multi-scale Dilated Convolution Network For Singer Identification

Abstract

Most singer identification methods are processed in the frequency domain, which potentially leads to information loss during the spectral transformation. In this paper, instead of the frequency domain, we propose an end-to-end architecture that addresses this problem in the waveform domain. An encoder based on Multi-scale Dilated Convolution Neural Networks (MDCNN) was introduced to generate wave embedding from the raw audio signal. Specifically, dilated convolution layers are used in the proposed method to enlarge the receptive field, aiming to extract song-level features. Furthermore, skip connection in the backbone network integrates the multi-resolution acoustic features learned by the stack of convolution layers. Then, the obtained wave embedding is passed into the following networks for singer identification. In experiments, the proposed method achieves comparable performance on the benchmark dataset of Artist20, which significantly improves related works.

Stats

• citations4
• S2 citations—
• github stars0
• HF likes0
• heat score5.24
• arxiv keyzhang2020mdcnn

Related papers

• D3net: Densely Connected Multidilated Densenet For Music Source Separation (2020)0.00
• Multi-band Multi-resolution Fully Convolutional Neural Networks For Singing Voice Separation (2019)5.84
• Boosting The Predictive Accurary Of Singer Identification Using Discrete Wavelet Transform For Feature Extraction (2021)0.00
• Mmdenselstm: An Efficient Combination Of Convolutional And Recurrent Neural Networks For Audio Source Separation (2018)15.28
• Multi-singer: Fast Multi-singer Singing Voice Vocoder With A Large-scale Corpus (2021)13.28
• Evolving Multi-resolution Pooling CNN For Monaural Singing Voice Separation (2020)9.03
• Unsupervised Singing Voice Conversion (2019)11.19
• Metasid: Singer Identification With Domain Adaptation For Metaverse (2022)7.50