The IQIYI System for Voice Conversion Challenge 2020

TL;DR

The paper introduces the IQIYI voice conversion system for the 2020 challenge, utilizing an end-to-end approach based on PPG, ASR features, and improved neural models to enhance voice naturalness and similarity.

Contribution

It presents a novel end-to-end voice conversion system combining PPG, improved prosody Tacotron, and LPCNet, achieving competitive results in the challenge.

Findings

Achieved high similarity and naturalness in voice conversion.

Performed well in objective and subjective evaluations.

Effective at 16k sampling rate for better quality.

Abstract

This paper presents the IQIYI voice conversion system (T24) for Voice Conversion 2020. In the competition, each target speaker has 70 sentences. We have built an end-to-end voice conversion system based on PPG. First, the ASR acoustic model calculates the BN feature, which represents the content-related information in the speech. Then the Mel feature is calculated through an improved prosody tacotron model. Finally, the Mel spectrum is converted to wav through an improved LPCNet. The evaluation results show that this system can achieve better voice conversion effects. In the case of using 16k rather than 24k sampling rate audio, the conversion result is relatively good in naturalness and similarity. Among them, our best results are in the similarity evaluation of the Task 2, the 2nd in the ASV-based objective evaluation and the 5th in the subjective evaluation.