采用STRAIGHT模型和深度信念网络的语音转换方法

提出一种将STRAIGHT模型和深度信念网络DBN相结合实现语音转换的方式。首先,通过STRAIGHT模型提取出源说话人和目标说话人的语音频谱参数,用提取的频谱参数分别训练两个DBN得到语音高阶空间的个性特征信息;然后,用人工神经网络ANN将两个具有高阶特征的空间连接并进行特征转换;最后,用基于目标说话人数据训练出的DBN来对转换后的特征信息进行逆处理得到语音频谱参数,并用STRAIGHT模型合成具有目标说话人个性化特征的语音。实验结果表明,采用此种方式获得的语音转换效果要比传统的采用GMM实现语音转换更好,转换后的语音音质和相似度与目标语音更接近。     

以CH563为核心的以太网语音通信终端设计

介绍了以嵌入式处理器CH563为核心,采用VS1053对语音信号进行压缩/解压缩处理,通过以太网进行数据交换的设计方案.在系统硬件设计方面,主要介绍了CH563与VS1053的性能参数及两者接口电路设计.在软件设计方面,主要介绍了以太网协议程序设计以及VS1053的操作方法,并给出具体的操作程序.最后介绍了该方案的优势,给出其在相关应用领域的研究方向.     

人机交互中的声源定位与增强方法

基于人机交互的实际应用场景,提出了一种交互目标声源三维定位与语音增强算法。该算法首先在广义相关法的基础上提出一种声达时延差的估计方法,通过由6个麦克风构成的平行均匀线阵接收模型,实现对目标声源的三维定位;然后在交互目标声源定位的基础上,通过调整时延波束叠加的权值,实现对目标语音的增强。仿真实验结果表明,文中提出的算法能够准确定位目标声源并对目标语音进行有效增强。在信噪比大于1.5 dB 的环境中,该算法可使目标声源达到98%以上定位精度,5 dB 以上信噪比改善,同时算法运算量小且易于硬件实现。     

多速率数字语音保密通信系统硬件设计

论文针对复杂的无线信道环境,设计实现了高质量、低误码率的语音保密通信系统.该系统完成一个发送通路和一个接收通路,可以完成一个完整的自环语音通信.该系统中,可供选择的数据编码速率为2000、3600、4000、4800、6400、7200、8000和9600bps.当外部干扰比较小时,可以选用较高的数据速率;当外部干扰较大时,可以选用较低的数据速率,以牺牲数据速率来提高语音通信的质量.     

基于语义搜索的实名呼叫系统设计

网络和通信技术的飞速发展,使传统的通信方式已经不能满足人们的需求。通信的人性化和智能化成为各个通信厂商追求的目标,但却始终无法摆脱号码这一枷锁。本文针对只能通过号码呼叫带来的通信不便问题,提出利用语义搜索这一未来互联网搜索技术实现实名呼叫的设计方案,并将语音搜索的思想融入其中,使通信变得更简单,实现通信的以人为本。     

ISD2560在门禁系统语音播报中的应用

设计并实现了由AT89S51和ISD2560组成的门禁系统语音播报系统。系统通过按键方式实现语音的分段录制,应用ISD2560信息检索工作模式以及按键控制工作模式,实现语音的选择播放,并且给出了该系统的硬件电路及一些主要的软件设计流程图。     

Triangulation in practice

Triangulation is the means by which an alternate perspective is used to validate, challenge or extend existing findings. It is frequently used when the field of study is difficult, demanding or contentious and presence research meets all of these criteria. We distinguish between the use of hard and soft triangulation—the former emphasising the challenging of findings, the latter being more confirmatory in character. Having reviewed a substantial number of presence papers, we conclude that strong triangulation is not widely used while soft triangulation is routinely employed. We demonstrate the usefulness of hard triangulation by contrasting an ontological analysis of in-ness with an empirical study of (computer) game playing. We conclude that presence research would be well served by the wider use of hard triangulation and for the reporting of anomalous and ill-fitting results.

Supporting the creative drive: investigating paralinguistic voice as a mode of interaction for artists with upper limb disabilities

For people with upper limb disabilities visual art is an important activity that allows for expression of individuality and independence. They show remarkable endurance, patience and determination to adapt their remaining capabilities to create visual art. There are significant advantages of digital technologies in assisting artists with upper limb disabilities. Paralinguistic voice recognition technologies have proven to be a particularly promising mode of interaction. Despite these benefits, technological support for people with upper limb disabilities to create visual art is scarce. This paper reports on a number of case studies of several artists with upper limb disabilities. These case studies illustrate the struggles they face to be creative and also show the significant advantages of digital technologies in assisting such artists. An investigation into people’s ability to use the volume of their voice to control cursor movement to create drawings on the screen is also reported. With motivation, training and practise, use of volume to control drawing tasks shows great promise. It is believed that paralinguistic voice has wider implications beyond assisting artists with upper limb disabilities, such as: an alternative mode of interaction for disabled people to perform tasks other than creating visual art, alternative mode of interaction for hands busy environments and as a voice training system for people with speech impairments.     

An improved method for voice pathology detection by means of a HMM-based feature space transformation

This paper presents new a feature transformation technique applied to improve the screening accuracy for the automatic detection of pathological voices. The statistical transformation is based on Hidden Markov Models, obtaining a transformation and classification stage simultaneously and adjusting the parameters of the model with a criterion that minimizes the classification error. The original feature vectors are built up using classic short-term noise parameters and mel-frequency cepstral coefficients. With respect to conventional approaches found in the literature of automatic detection of pathological voices, the proposed feature space transformation technique demonstrates a significant improvement of the performance with no addition of new features to the original input space. In view of the results, it is expected that this technique could provide good results in other areas such as speaker verification and/or identification.     

Effectiveness of MMORPG‐based instruction in elementary English education in Korea

This study investigated the effectiveness of massive multiplayer online role‐playing game (MMORPG)‐based (massive multiplayer online role‐playing game) instruction in elementary English education. The effectiveness of the MMORPG program was compared with face‐to‐face instruction and the independent variables (gender, prior knowledge, motivation for learning, self‐directed learning skills, computer skills, game skills, computer capacity, network speed, and computer accessibility) were examined to see how accurately achievement was predicted in MMORPG instruction. The results indicated that students studying English utilizing online role‐playing games showed higher scores in areas of listening, reading, and writing than those who attended face‐to‐face instruction classes. It was also found that prior knowledge, motivation for learning, and network speed were factors affecting achievement in English learning. These findings suggest that MMORPGs can play an important role in improving English communicative skills.