Stage-Wise and Prior-Aware Neural Speech Phase Prediction

TL;DR

This paper introduces a two-stage neural network model for speech phase prediction that leverages prior phase information and adversarial training to improve accuracy and efficiency over existing methods.

Contribution

The paper presents a novel stage-wise, prior-aware neural model with adversarial training and a new loss function for improved speech phase prediction.

Findings

Achieves higher phase prediction accuracy than no-prior methods.

Does not require iterative estimation, increasing efficiency.

Outperforms existing neural and iterative algorithms in experiments.

Abstract

This paper proposes a novel Stage-wise and Prior-aware Neural Speech Phase Prediction (SP-NSPP) model, which predicts the phase spectrum from input amplitude spectrum by two-stage neural networks. In the initial prior-construction stage, we preliminarily predict a rough prior phase spectrum from the amplitude spectrum. The subsequent refinement stage transforms the amplitude spectrum into a refined high-quality phase spectrum conditioned on the prior phase. Networks in both stages use ConvNeXt v2 blocks as the backbone and adopt adversarial training by innovatively introducing a phase spectrum discriminator (PSD). To further improve the continuity of the refined phase, we also incorporate a time-frequency integrated difference (TFID) loss in the refinement stage. Experimental results confirm that, compared to neural network-based no-prior phase prediction methods, the proposed SP-NSPP achieves higher phase prediction accuracy, thanks to introducing the coarse phase priors and diverse training criteria. Compared to iterative phase estimation algorithms, our proposed SP-NSPP does not require multiple rounds of staged iterations, resulting in higher generation efficiency.