Speaker anonymization by pitch shifting based on time-scale modification
Speaker anonymization by pitch shifting based on time-scale modification
复制标题
DOI：
10.21437/spsc.2022-7
复制
发表时间：
2022-09
期刊：
影响因子：
--
通讯作者：
Candy Olivia Mawalim;S. Okada;M. Unoki
中科院分区：
文献类型：
--
作者：
Candy Olivia Mawalim;S. Okada;M. Unoki
数字技术中语音的使用越来越多，这引起了隐私问题，因为语音包含生物特征信息。已经提出了几种处理这个问题的方法，包括说话人匿名化或去识别。说话人匿名化的目的是在保留语音的其他属性（包括语言信息）的同时，抑制个人身份信息（PII）。在这项研究中，我们利用时间尺度修改（TSM）语音信号处理的说话人匿名。语音信号处理方法比最先进的基于x向量的说话人匿名化方法复杂得多，因为它不需要训练过程。我们提出了匿名化方法，使用两个主要类别的TSM，同步的语音添加（SOLA）为基础的算法和相位声码器为基础的TSM（PV-TSM）。为了评估我们提出的方法，我们利用标准的客观评估中引入的语音隐私的挑战。结果表明，我们基于PV-TSM的方法比基线系统更好地平衡了隐私和实用性指标，特别是在匿名注册和匿名试验（a-a）中使用自动说话人验证（ASV）系统进行评估时。此外，我们的方法优于基于x向量的说话人方法，后者在训练过程复杂、a-a场景中隐私性低以及语音独特性低方面存在局限性。
The increasing usage of speech in digital technology raises a privacy issue because speech contains biometric information. Several methods of dealing with this issue have been proposed, including speaker anonymization or de-identification. Speaker anonymization aims to suppress personally identifiable information (PII) while keeping the other speech properties, including linguistic information. In this study, we utilize time-scale modification (TSM) speech signal processing for speaker anonymization. Speech signal processing approaches are significantly less complex than the state-of-the-art x-vector-based speaker anonymization method because it does not require a training process. We propose anonymization methods using two major categories of TSM, synchronous overlap-add (SOLA)- based algorithm and phase vocoder-based TSM (PV-TSM). For evaluating our proposed methods, we utilize the standard objec-tive evaluation introduced in the VoicePrivacy challenge. The results show that our method based on the PV-TSM balances privacy and utility metrics better than baseline systems, especially when evaluating with an automatic speaker verification (ASV) system in anonymized enrollment and anonymized trials (a-a). Further, our method outperformed the x-vector-based speaker method, which has limitations in its complex training process, low privacy in an a-a scenario, and low voice distinctiveness.