Phase reconstruction based on recurrent phase unwrapping with deep neural networks

TL;DR

This paper introduces a two-stage deep neural network method for phase reconstruction that estimates phase derivatives to improve accuracy, utilizing recurrent phase unwrapping to enhance audio synthesis applications.

Contribution

The paper proposes a novel DNN-based two-stage phase reconstruction method that estimates phase derivatives and employs recurrent phase unwrapping, addressing phase sensitivity issues.

Findings

Outperforms direct phase estimation methods

Effectively estimates phase derivatives for improved accuracy

Demonstrates robustness in phase reconstruction tasks

Abstract

Phase reconstruction, which estimates phase from a given amplitude spectrogram, is an active research field in acoustical signal processing with many applications including audio synthesis. To take advantage of rich knowledge from data, several studies presented deep neural network (DNN)--based phase reconstruction methods. However, the training of a DNN for phase reconstruction is not an easy task because phase is sensitive to the shift of a waveform. To overcome this problem, we propose a DNN-based two-stage phase reconstruction method. In the proposed method, DNNs estimate phase derivatives instead of phase itself, which allows us to avoid the sensitivity problem. Then, phase is recursively estimated based on the estimated derivatives, which is named recurrent phase unwrapping (RPU). The experimental results confirm that the proposed method outperformed the direct phase estimation by a DNN.