PSVRF: Learning to restore Pitch-Shifted Voice without reference

TL;DR

This paper introduces PSVRF, a no-reference method for restoring pitch-shifted voices, significantly improving ASV system robustness against pitch-scaling attacks without needing the original voice as a reference.

Contribution

PSVRF is the first no-reference approach that effectively restores pitch-shifted voices, outperforming existing reference-based methods in quality and robustness.

Findings

PSVRF successfully restores pitch-shifted voices across various techniques.

It enhances the robustness of ASV systems against pitch-scaling attacks.

PSVRF outperforms state-of-the-art reference-based approaches.

Abstract

Pitch scaling algorithms have a significant impact on the security of Automatic Speaker Verification (ASV) systems. Although numerous anti-spoofing algorithms have been proposed to identify the pitch-shifted voice and even restore it to the original version, they either have poor performance or require the original voice as a reference, limiting the prospects of applications. In this paper, we propose a no-reference approach termed PSVRF$^1$ for high-quality restoration of pitch-shifted voice. Experiments on AISHELL-1 and AISHELL-3 demonstrate that PSVRF can restore the voice disguised by various pitch-scaling techniques, which obviously enhances the robustness of ASV systems to pitch-scaling attacks. Furthermore, the performance of PSVRF even surpasses that of the state-of-the-art reference-based approach.