MaD TwinNet: Masker-Denoiser Architecture with Twin Networks for Monaural Sound Source Separation

被引：0

作者：

Drossos, Konstantinos ^{[1
]}

Mimilakis, Stylianos Ioannis ^{[2
]}

Serdyuk, Dmitriy ^{[3
]}

Schuller, Gerald ^{[2
]}

Virtanen, Tuomas ^{[1
]}

Bengio, Yoshua ^{[3
]}

机构：

[1] Tampere Univ Technol, Tampere, Finland

[2] Tech Univ Ilmenau, Fraunhofer IDMT, Ilmenau, Germany

[3] Univ Montreal, MILA, Montreal, PQ, Canada

来源：

2018 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN) | 2018年

基金：

欧盟地平线“2020”; 欧洲研究理事会; 加拿大自然科学与工程研究理事会;

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Monaural singing voice separation task focuses on the prediction of the singing voice from a single channel music mixture signal. Current state of the art (SOTA) results in monaural singing voice separation are obtained with deep learning based methods. In this work we present a novel recurrent neural approach that learns long-term temporal patterns and structures of a musical piece. We build upon the recently proposed Masker-Denoiser (MaD) architecture and we enhance it with the Twin Networks, a technique to regularize a recurrent generative network using a backward running copy of the network. We evaluate our method using the Demixing Secret Dataset and we obtain an increment to signal-to-distortion ratio (SDR) of 0.37 dB and to signal-to-interference ratio (SIR) of 0.23 dB, compared to previous SOTA results.

引用

页数：8