全複數深度遞迴類神經網路應用於歌曲人聲分離;Complex-Valued Deep Recurrent Neural Network for Singing Voice Separation

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/74713

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/74713

Title:	全複數深度遞迴類神經網路應用於歌曲人聲分離;Complex-Valued Deep Recurrent Neural Network for Singing Voice Separation
Authors:	俞果;Yu, Kuo
Contributors:	資訊工程學系
Keywords:	深度類神經網路;歌唱人聲分離;相位資訊;Deep Neural Network;Singing Voice Separation;Phase Informaiton
Date:	2017-08-14
Issue Date:	2017-10-27 14:37:03 (UTC+8)
Publisher:	國立中央大學
Abstract:	深度類神經網路(DNN)在多媒體訊號處理的領域中有不凡的表現，但大部分的基於深度類神經網路的作法都是在處理實數資料，只有少數設計成能處理複數資料，即便複數資料在多媒體的領域佔有重要的地位，因此本論文提出全複數深度遞迴類神經網路(C-DRNN)的架構來處理歌曲人聲分離，本架構可以直接處理短時傅立葉轉換(STFT)出來的複數資料，並且本架構的權重以及激發函數等都是以複數計算。本論文的目標為從歌曲中分離人聲與樂器，在倒傳遞時使用複數微分成本函數，進而得到複數梯度，本架構也對輸出層做了改進，加入了複數比例遮罩以確保最後估計的輸出不會超過輸入的數值，並且在訓練網路時多加了鑑別項以增加網路的計算能力。最後，本論文提出的方法使用MIR-1K資料庫實驗歌曲人聲分離的能力，實驗結果顯示本方法較其他深度類神經網路表現更加優秀。;Deep neural networks (DNN) have performed impressively in the processing of multimedia signals. Most DNN-based approaches were developed to handle real-valued data; very few have been designed for complex-valued data, despite their being essential for processing various types of multimedia signal. Accordingly, this work presents a complex-valued deep recurrent neural network (C-DRNN) for singing voice separation. The C-DRNN operates on the complex-valued short-time discrete Fourier transform (STFT) domain. A key aspect of the C-DRNN is that the activations and weights are complex-valued. The goal herein is to reconstruct the singing voice and the background music from a mixed signal. For error back-propagation, CR-calculus is utilized to calculate the complex-valued gradients of the objective function. To reinforce model regularity, two constraints are incorporated into the cost function of the C-DRNN. The first is an additional masking layer that ensures the sum of separated sources equals the input mixture. The second is a discriminative term that preserves the mutual difference between two separated sources. Finally, the proposed method is evaluated using the MIR-1K dataset and a singing voice separation task. Experimental results demonstrate that the proposed method outperforms the state-of-the-art DNN-based methods.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	282	View/Open

社群 sharing

Loading...