Pseudo-Cepstrum: Pitch Modification for Mel-Based Neural Vocoders

TL;DR

This paper presents a cepstrum-based pitch modification technique compatible with any mel-based neural vocoder, enabling pitch shifting without retraining or model modifications by directly manipulating the cepstral domain.

Contribution

It introduces a novel, model-agnostic pitch modification method using cepstral domain manipulation that works with any mel-spectrogram based neural vocoder.

Findings

Effective pitch shifting demonstrated on various neural vocoders.

Method achieves high-quality speech with minimal artifacts.

Outperforms traditional pitch modification techniques in experiments.

Abstract

This paper introduces a cepstrum-based pitch modification method that can be applied to any mel-spectrogram representation. As a result, this method is compatible with any mel-based vocoder without requiring any additional training or changes to the model. This is achieved by directly modifying the cepstrum feature space in order to shift the harmonic structure to the desired target. The spectrogram magnitude is computed via the pseudo-inverse mel transform, then converted to the cepstrum by applying DCT. In this domain, the cepstral peak is shifted without having to estimate its position and the modified mel is recomputed by applying IDCT and mel-filterbank. These pitch-shifted mel-spectrogram features can be converted to speech with any compatible vocoder. The proposed method is validated experimentally with objective and subjective metrics on various state-of-the-art neural vocoders as well as in comparison with traditional pitch modification methods.