NTU-NPU System for Voice Privacy 2024 Challenge

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kuzmin, Nikita, Luong, Hieu-Thi, Yao, Jixun, Xie, Lei, Lee, Kong Aik, Chng, Eng Siong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914968932188160
author Kuzmin, Nikita
Luong, Hieu-Thi
Yao, Jixun
Xie, Lei
Lee, Kong Aik
Chng, Eng Siong
author_facet Kuzmin, Nikita
Luong, Hieu-Thi
Yao, Jixun
Xie, Lei
Lee, Kong Aik
Chng, Eng Siong
contents In this work, we describe our submissions for the Voice Privacy Challenge 2024. Rather than proposing a novel speech anonymization system, we enhance the provided baselines to meet all required conditions and improve evaluated metrics. Specifically, we implement emotion embedding and experiment with WavLM and ECAPA2 speaker embedders for the B3 baseline. Additionally, we compare different speaker and prosody anonymization techniques. Furthermore, we introduce Mean Reversion F0 for B5, which helps to enhance privacy without a loss in utility. Finally, we explore disentanglement models, namely $β$-VAE and NaturalSpeech3 FACodec.
format Preprint
id arxiv_https___arxiv_org_abs_2410_02371
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle NTU-NPU System for Voice Privacy 2024 Challenge
Kuzmin, Nikita
Luong, Hieu-Thi
Yao, Jixun
Xie, Lei
Lee, Kong Aik
Chng, Eng Siong
Audio and Speech Processing
Artificial Intelligence
In this work, we describe our submissions for the Voice Privacy Challenge 2024. Rather than proposing a novel speech anonymization system, we enhance the provided baselines to meet all required conditions and improve evaluated metrics. Specifically, we implement emotion embedding and experiment with WavLM and ECAPA2 speaker embedders for the B3 baseline. Additionally, we compare different speaker and prosody anonymization techniques. Furthermore, we introduce Mean Reversion F0 for B5, which helps to enhance privacy without a loss in utility. Finally, we explore disentanglement models, namely $β$-VAE and NaturalSpeech3 FACodec.
title NTU-NPU System for Voice Privacy 2024 Challenge
topic Audio and Speech Processing
Artificial Intelligence
url https://arxiv.org/abs/2410.02371