权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Development and Construct of Speech Privacy Protection System Using Masking Effect

利用掩蔽效应的语音隐私保护系统的开发与构建

基本信息

批准号：
15500138
负责人：
YAMAGUCHI Shizuma
金额：
$ 2.3万
依托单位：
Yamaguchi University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2003
资助国家：
日本
起止时间：
2003 至 2005
项目状态：
已结题

项目摘要

Recently the importance of achieving speech privacy in open spaces has been emphasized. In this study, a system for achieving speech privacy using a masking noise was developed. We adopted a meaningless steady noise to mask speeches that has various power spectra, and closed headphones/throat microphones to deliver clear speech even in high noise environments.1.The most effective index for achieving a satisfactory level of speech privacy is selected by evaluating various indices. From the results, spectral distance was selected as the best index for evaluating speech privacy.2.The spectrum of speech differs between individuals, so using masking noise corresponding to individual voice spectra was effective in minimizing annoyance. From this viewpoint, a device was constructed to generate noise with the same spectrum form as a voice. Psychological experiments were conducted to consider the validity of the device constructed.3.The above system had some problems : occasionally it radiated too much masking noise, and voice communication quality was slightly poor because the frequency characteristics of the throat microphones were not flat. To consider the effectiveness of reducing the noisiness and annoyance by using masking noise which takes background noise into account, and the effectiveness of improving audibility and listening score by correcting frequency characteristic of throat microphones, psychological experiments were conducted.4.Psychological experiments were conducted to evaluate the developed compact/portable/light-wait system. The results confirmed the usefulness of the developed system.

近年来，在开放空间中实现语音隐私的重要性得到了强调。在这项研究中，开发了一种利用掩蔽噪声实现语音隐私的系统。我们采用了无意义的稳定噪声来掩盖具有不同功率谱的语音，并闭合耳机/喉咙麦克风以在高噪声环境下提供清晰的语音。1.通过对各种指标的评估来选择最有效的指标来实现令人满意的语音隐私水平。结果表明，频谱距离作为评价语音隐私的最佳指标。2.语音频谱因个体不同而不同，因此使用与个体语音频谱相对应的掩蔽噪声可以有效地减少烦扰。从这一观点来看，一种设备被构造为产生与语音具有相同频谱形式的噪声。通过心理学实验验证了所构建装置的有效性。3.上述系统存在一些问题：偶尔会辐射过多的掩蔽噪声，并且由于喉部麦克风的频率特性不平坦，语音通信质量略差。为了考察利用考虑背景噪声的掩蔽噪声来降低噪声和烦扰的有效性，以及通过校正喉部麦克风的频率特性来提高听觉和听力分数的有效性，进行了心理学实验。4.对所开发的紧凑型/便携/光等待系统进行了心理学实验评估。结果证实了所开发系统的实用性。