A Recurrent Encoder-Decoder Approach with Skip-filtering Connections for Monaural Singing Voice Separation

TL;DR

This paper introduces a recurrent encoder-decoder model with skip-filtering connections that directly learns time-frequency masks for monaural singing voice separation, achieving improved signal quality over previous mask-approximation methods.

Contribution

The proposed method directly learns time-frequency masks using recurrent neural networks with skip-filtering connections, simplifying the process and improving separation quality.

Findings

Achieves comparable results to complex deep learning methods.

Increases signal-to-distortion ratio by an average of 3.8 dB.

Effective for singing voice separation from monaural recordings.

Abstract

The objective of deep learning methods based on encoder-decoder architectures for music source separation is to approximate either ideal time-frequency masks or spectral representations of the target music source(s). The spectral representations are then used to derive time-frequency masks. In this work we introduce a method to directly learn time-frequency masks from an observed mixture magnitude spectrum. We employ recurrent neural networks and train them using prior knowledge only for the magnitude spectrum of the target source. To assess the performance of the proposed method, we focus on the task of singing voice separation. The results from an objective evaluation show that our proposed method provides comparable results to deep learning based methods which operate over complicated signal representations. Compared to previous methods that approximate time-frequency masks, our method has increased performance of signal to distortion ratio by an average of 3.8 dB.