Frequency Selectivity Behaviour in the Auditory Midbrain： Implications of Model Study

By numerical simulations on frequency dependence of the spiking threshold, i.e. on the critical amplitude of periodic stimulus, for a neuron to fire, we find that bushy cells in the cochlear nuclear exhibit frequency selec- tivity behaviour. However, the selective frequency band of a bushy cell is far away from that of the preferred spectral range in human and mammal auditory perception. The mechanism underlying this neural activity is also discussed. Further studies show that the ion channel densities have little impact on the selective frequency band of bushy cells. These findings suggest that the neuronal behaviour of frequency selectivity in bushy cells at both the single cell and population levels may be not functionally relevant to frequency discrimination. Our results may reveal a neural hint to the reconsideration on the busily cell functional role in auditory information processing of sound frequency.     

Auditory place theory and frequency difference limen

It has been a barrier that the place code is far too coarse a mechanism to account for the finest frequency difference limen for place theory of hearing since it was proposed in 19th century. A place correlation model, which takes the energy distribution of a pure tone in neighboring bands of auditory filters into full account, was presented in this paper. The model based on the place theory and some experimental results of the psychophysical tuning curves of hearing can explain the finest difference limen for frequency (about 0.02 or 0.3% at 1000 Hz) easily. Using a standard 1/3 octave filter bank of which the relationship between the frequency of a input pure tone apart from the centre frequency of K-th filter band, Δf, and the output intensity difference between K-th and (K l)-th filters,ΔE, was established in order to show the fine frequency detection ability of the filter bank. This model can also be used to abstract the fundamental frequency of speech and to measure the frequency of pure tone precisely.     

幅度调制信息对乐器识别的影响

通过乐器音自动识别实验研究了幅度调制(amplitude modulation,AM)信息对于人类乐器识别的影响。具体步骤为:依据听觉模型,提取乐器音信号中若干频带中的AM信息,再基于所得到的AM信息计算统计学特征,采用逐对比较的支持向量机法进行乐器音的机器识别。采用了5种分频带数目(2,4,8,16和32)和4种AM计算方法。结果表明,频带数的增加有助于识别效果的提高,但从16频带到32频带效果趋近平稳;不同的AM提取方法也会对识别结果产生影响,其中解析信号法产生的AM信息提供了最好的乐器识别效果。分析发现自动识别结果高于采用相似的AM信息的人类识别结果。该自动识别系统为人工耳蜗或声码器仿真声模型的乐器识别提供了一个计算模型,对人工耳蜗乐器识别实验和训练具有参考价值。      

南下访“淇淇”

介绍甸独遥长江淡水白暨豚“淇淇”及其同种的发声和听觉的特性,包括它们所发射超声信号（的答声）和可听声信号（哨叫声）以及它们的听觉区域,也介绍了白暨豚当前频临灭绝的处境 。     

Blind estimation of shallow water acoustic channel

This paper proposed a method for blind estimation of underwater channels in shallow water environment by using received data at a single hydrophone or from single beam. First,the received signal is used for source signal reconstruction by means of signal-dependent TF(Time-Frequency)distribution,in association with instantaneous frequency estimation and TF inversion.Then the shallow-water channel estimation is achieved via WRELAX technique by use of the received signal and the estimated source signal.Finally,the results of numerical simulation and experimental test from real data taken in South China Sea trial have proved satisfactory.It is shown that the proposed method is useful for underwater channel estimation.     

听觉基础研究的若干问题展望(2)

三、心理声学 当前心理声学研究的主要趋向是从以往纯音听觉的单因素研究转向具有频谱时间表征的复合声的多维研究。这个转变和生理声学的发展情况恰成有趣的对比。如上节所述,生理声学的近期进展是对Helmholtz为代表的传统耳蜗观念的突破。与此相反,心理声学的前进却不能不回到Helmholtz在其名著“乐音感觉:音乐理论的生理学基础”一书中提出的一个基本课题,即对音乐、言语这类复合声的知觉。自然,这并不是说心理声学应该迳直去研究音乐和言语,去取代音乐声学和言语声学。在     

基于样本熵的听觉神经锁相机理的实验分析

锁相是指系统的响应与周期性刺激的特定相位同步的物理现象. 听觉神经的锁相对揭示人的听觉认知基本的神经机理及改善听觉感知有重要意义. 然而, 现有研究主要集中于心理物理方法和幅度谱分析, 不能有效区分包络响应和时域细节结构响应, 不能直观反映神经锁相. 本文主要利用拔靴法和离散傅里叶变换, 提出了基于样本熵的时域细节结构频率跟随响应(temporal-fine-structure-related frequency following response, FFR_T)的神经锁相值(phase locking value, PLV)计算方法, 用于分析神经物理实验数据. 两个脑电实验结果表明: FFR_T的PLV样本熵显著大于包络相关频率跟随响应(envelope-related frequency following response, FFR_E)的PLV, 且二者正交独立, 新方法能有效地分别反映听觉系统对包络和时间细节结构的锁相机理; 基频处的响应主要来源于FFR_E的锁相; 基频处, 不可分辨谐波成分包络的锁相能力优于对可分辨谐波; 基频缺失时, 畸变产物是不同的听觉神经通路的FFR_E的混合; 谐波处, FFR_E 集中于低频, FFR_T则集中于中、高频; 听觉神经元锁相能力与声源的频率可分辨性相关. FFR_T的PLV方法克服了现有FFR分析的局限性, 可用于深入研究听觉神经机理.     

声学随机共振实验

在特定的非线性系统中,噪声的存在能够增强信号的响应, 这种现象为随机共振. 本文实验通过观察噪声对听觉阈限的影响,使学生了解声学中的随机共振现象.     

人类对超声的听觉感知及应用前景

本文对人类通过骨传导所产生的超声听觉感知，就以下四方面进行介绍：１．感知的方法、频率和强度范围；２．感知机理；３．可能的应用；４．值得研究的问题。     

QSC-1、QSC-2型超声驱鼠器

老鼠为人类之大敌.历来人类控制鼠害是用捕杀和毒杀的方法,收到一定效果.但由于老鼠生性狡猾、行动敏捷、生存力强、繁殖率高,鼠害仍然普遍存在.在一些特别需要防止老鼠的场所,如图书资料库、计算机房、电器控制室、药品库以及其它仓库、饭馆、食堂等,除捕杀外,驱鼠也是一种保护物资、防止事故、减少