Most recent speech privacy efforts have focused on anonymizing acoustic
speaker attributes but there has not been as much research into protecting
information from speech content. We introduce a toy problem that explores an
emerging type of privacy called "content masking" which conceals selected words
and phrases in speech. In our efforts to define this problem space, we evaluate
an introductory baseline masking technique based on modifying sequences of
discrete phone representations (phone codes) produced from a pre-trained
vector-quantized variational autoencoder (VQ-VAE) and re-synthesized using
WaveRNN. We investigate three different masking locations and three types of
masking strategies: noise substitution, word deletion, and phone sequence
reversal. Our work attempts to characterize how masking affects two downstream
tasks: automatic speech recognition (ASR) and automatic speaker verification
(ASV). We observe how the different masks types and locations impact these
downstream tasks and discuss how these issues may influence privacy goals.

最近的语音隐私研究主要集中在匿名化声学说话者属性，但对于保护语音内容的信息研究较少。本文引入了一个玩具问题，探讨了一种被称为 “内容隐藏” 的新型隐私保护方法，该方法隐藏了语音中的选定单词和短语。通过改变预训练的向量量化变分自编码器（VQ-VAE）生成的离散电话表示（电话代码）序列并通过 WaveRNN 重新合成的方法，我们评估了一个基准掩码技术。我们对三种不同的掩码位置和三种掩码策略进行了研究：噪声替换、单词删除和电话序列反转。我们着重研究了掩码对自动语音识别（ASR）和自动说话者验证（ASV）这两个后续任务的影响，并讨论了如何影响隐私目标的这些问题。