基于听觉模型的耳语音的声韵切分

本文分析了耳语音的特点，并根据生理声学及心理声学的基本理论与实验资料，提出了一种利用听觉模型来进行耳语音声韵切分的方法。这种适用于耳语音声韵切分的听觉感知模型主要分为四个层次：耳蜗对声音频率的分解机理；听觉系统的时域和频域非线性变化；中枢神经系统的侧抑制机理。这种模型能反映在噪声环境下人对低能量语音的听觉感知特性，因而适于耳语音识别，在耳语音声韵母切分实验中得到了满意的结果。     

模型与特征混合补偿法及其在耳语说话人识别中的应用

为了提高信道差异下短时耳语说话人的识别率,提出了一种在模型域和特征域进行混合补偿的方法。该方法首先在模型训练阶段以联合因子分析法为基础,通过估计训练语音的说话人空间和信道空间,提取出说话人因子,消除信道因子,其次在测试阶段,将测试语音的信道因子映射到特征空间,实施特征补偿,从而在模型和特征两方面去除信道信息,提高识别率。实验结果显示,在三种不同的信道训练环境下,混合补偿法都取得了相似的识别率,且新方法对短时耳语音的测试效果要优于联合因子分析法。      

汉语耳语音孤立字识别研究

耳语音识别有着广泛的应用前景,是一个全新的课题.但是由于耳语音本身的特点,如声级低、没有基频等,给耳语音识别研究带来了困难.本文根据耳语音信号发音模型,结合耳语音的声学特性,建立了一个汉语耳语音孤立字识别系统.由于耳语音信噪比低,必须对其进行语音增强处理,同时在识别系统中应用声调信息提高了识别性能.实验结果说明了MFCC结合幅值包络可作为汉语耳语音自动识别的特征参数,在小字库内用HMM模型识别得出的识别率为90.4%.     

倒谱本征空间结构化高斯混合模型语音转换方法

针对非平行语料非联合训练条件下的语音转换,提出一种基于倒谱本征空间结构化高斯混合模型的方法。提取说话人语音倒谱特征参数之后,根据其散布矩阵计算本征向量构造倒谱本征空间并训练结构化高斯混合模型SGMM-ES(Structured Gaussian Mixture Model in Eigen Space)。源和目标说话人各自独立训练的SGMM-ES根据全局声学结构AUS(Acoustical Universal Structure)原理进行匹配对准,最终得到基于倒谱本征空间的短时谱转换函数。实验结果表明,转换语音的目标说话人平均识别率达到95.25%,平均谱失真度为1.25,相对基于原始倒谱特征空间的SGMM方法分别提高了0.8%和7.3%,而ABX和MOS测评表明转换性能非常接近于传统平行语料方法。这一结果说明采用倒谱本征空间结构化高斯混合模型进行非平行语料条件下的语音转换是有效的。      

融合梅尔谱增强与特征解耦的噪声鲁棒语音转换

提出了一种融合梅尔谱增强与特征解耦的噪声鲁棒语音转换模型,即MENR-VC模型。该模型采用3个编码器提取语音内容、基频和说话人身份矢量特征,并引入互信息作为相关性度量指标,通过最小化互信息进行矢量特征解耦,实现对说话人身份的转换。为了改善含噪语音的频谱质量,模型使用深度复数循环卷积网络对含噪梅尔谱进行增强,并将其作为说话人编码器的输入;同时,在训练过程中,引入梅尔谱增强损失函数对模型整体损失函数进行了改进。仿真实验结果表明,与同类最优的噪声鲁棒语音转换方法相比,所提模型得到的转换语音在语音自然度和说话人相似度的平均意见得分方面,分别提高了0.12和0.07。解决了语音转换模型在使用含噪语音进行训练时,会导致深度神经网络训练过程难以收敛,转换语音质量大幅下降的问题。     

感知听觉场景分析的说话人识别

针对低信噪比说话人识别中缺失数据特征方法鲁棒性下降的问题,提出了一种采用感知听觉场景分析的缺失数据特征提取方法。首先求取语音的缺失数据特征谱,并由语音的感知特性求出感知特性的语音含量。含噪语音经过感知特性的语音增强和对其语谱的二维增强后求解出语音的分布,联合感知特性语音含量和缺失强度参数提取出感知听觉因子。再结合缺失数据特征谱把特征的提取过程分解为不同听觉场景进行区分地分析和处理,以增强说话人识别系统的鲁棒性能。实验结果表明,在-10 dB到10 dB的低信噪比环境下,对于4种不同的噪声,提出的方法比5种对比方法的鲁棒性均有提高,平均识别率分别提高26.0%,19.6%,12.7%,4.6%和6.5%。论文提出的方法,是一种在时-频域中寻找语音鲁棒特征的方法,更适合于低信噪比环境下的说话人识别。      

采用低维特征映射的耳语音向正常音转换

在将耳语音转换为正常音时,为了研究降维后语音特征对耳语音转换的影响,分别对耳语音和正常音谱包络进行自适应编码以提取耳语音和正常音的低维特征,然后使用BP网络建立耳语音和正常音低维谱包络特征之间的映射关系以及正常音基频和耳语音低维谱包络特征之间的关系。转换时,根据耳语音低维谱包络特征获得对应正常音的低维谱包络特征和基频,对低维谱包络特征进行解码后获得对应的正常音谱包络。实验结果表明,采用此方法转换后的语音与正常音之间的倒谱距离相比高斯混合模型方法下降了10%,转换后语音的自然度和可懂度都有所提高。      

基于听觉模型的话者特征参数提取及其在噪声背景下的话者辨识

本文基于人耳听觉模型提出了一种鲁棒性的话者特征参数提取方法。该种方法中,首先由Gamma tone听觉滤波器组和Meddis内耳毛细胞发放模型获得表征听觉神经活动特性的听觉相关图。由听觉神经脉冲发放的锁相特性和双声抑制特性,我们将听觉相关图每个频带中的幅值最大频率分量作为表征当前频带特性的特征参量,于是所有频带的特征参量便构成了表征当前语音段特性的特征矢量;我们采用DCT交换进一步消除各个特征参量之间的相关性,压缩特征矢量的维数。有效性试验表明,该种特征矢量基本上反映了输入语音的谱包络特性;抗噪声性能实验表明,在高斯白噪声和汽车噪声干扰下,这种特征参数比LPCC和MFCC有较小的相对失真;基于矢量量化的文本无关话者辨识表明,对于三种类型的噪声干扰该种特征参数在低信噪比下都获得了较好的识别结果。     

低信噪比下采用感知语谱结构边界参数的语音端点检测算法

提出了一种采用感知语谱结构边界参数(PSSB)的语音端点检测算法,用于在低信噪比环境下的语音信号预处理。在对含噪语音进行基于听觉感知特性的语音增强之后,针对语音信号的连续分布特性与残留噪声的随机分布特性之间的不同点,对增强后语音的时-频语谱进行二维增强,从而进一步突出连续分布的纯净语音的语谱结构。通过对增强后语音语谱结构的二维边界检测,提出PSSB参数,并用于端点检测。实验结果表明,在白噪声-10 dB到10 dB的各种信噪比环境下,采用PSSB参数的端点检测算法,相对于其它端点检测算法,更有效地检测出语音的端点。在-10 dB的极低信噪比下,提出的方法仍然有75.2%的正确率。采用PSSB参数的端点检测算法,更适合于低信噪比白噪声环境下的语音端点检测。      

采用扩展型双线性变换法将耳语音转换为正常语音的研究

提出了一种采用扩展型双线性变换将耳语音转换为正常语音的方法。根据耳语音在不同频段的共振峰偏移程度不同,将耳语音的频谱进行分段处理,在此基础上建立耳语音转换为正常语音的转换函数。由于耳语音在各频段相对于正常语音非线性偏移,在双线性变换函数中引入扩展因子,使其对频谱的非线性偏移与对共振峰带宽的压缩更加符合耳语音转换为正常语音的实际转换需求,有效减小了转换语音与正常语音的谱失真距离。实验结果表明,本文的转换语音在音质和可懂度上均得到了有效提高。      

Whispered speaker identification based on feature and model hybrid compensation

In order to increase short time whispered speaker recognition rate in variable channel conditions,the hybrid compensation in model and feature domains was proposed.This method is based on joint factor analysis in training model stage.It extracts speaker factor and eliminates channel factor by estimating training speech speaker and channel spaces.Then in the test stage,the test speech channel factor is projected into feature space to engage in feature compensation,so it can remove channel information both in model and feature domains in order to improve recognition rate.The experiment result shows that the hybrid compensation can obtain the similar recognition rate in the three different training channel conditions and this method is more effective than joint factor analysis in the test of short whispered speech.     

基于多带解调分析和瞬时频率估计的耳语音话者识别

为了改善耳语音话者识别的稳健性,提出了一种基于调幅-调频(AM-FM)模型的耳语音特征参数,瞬时频率估计(IFE)。根据语音产生的共振峰调制理论,采用多带解调分析(MDA)获得语音的瞬时包络和频率;然后根据包络幅度和频率的加权估计,得到语音的特征IFE来描绘语音的频率结构。将该特征用于耳语话者识别并和传统的Mel倒谱系数(MFCC)进行了比较。实验结果表明,随着测试人数的增加,IFE的识别效果略好于MFCC;在测试信道改变的情况下,与MFCC相比IFE的稳健性得到了有效的提高。      

Study of the formant and duration in Chinese whispered vowel speech

Study on the acoustical characteristic is important to speech and speaker recognition in Chinese whispered speech. In this paper, the characteristics of whispered speech are introduced and the acoustical characteristics in Chinese whispered speech are discussed. There is no fundamental frequency in the whispered speech, so other characteristics such as the duration and frequency of formant are extracted and analyzed. From experiments with six simple Chinese whispered vowels, it is proved that the duration and the frequency of formant can be used as the main acoustical characteristics in the Chinese whispered recognition.     

汉语耳语标准频谱的测量与计算

提出了与GB7348-87《耳语标准频谱》不同的汉语耳语功率谱密度级随频率的变化关系。在消声室中测量以提高测量信噪比,使用实时分析仪测量单个人耳语发音的长期声压频谱,并且对每个人的长期声压频谱做自归一化,通过数学方法将多个样本"混录",计算出汉语耳语的功率谱密度级。汉语耳语标准频谱的测量和计算结果可为一切产生、传输、接收和处理汉语耳语信号的系统及电声器件的设计提供依据。      

Speaker recognition with temporal cues in acoustic and electric hearing

Natural spoken language processing includes not only speech recognition but also identification of the speaker's gender, age, emotional, and social status. Our purpose in this study is to evaluate whether temporal cues are sufficient to support both speech and speaker recognition. Ten cochlear-implant and six normal-hearing subjects were presented with vowel tokens spoken by three men, three women, two boys, and two girls. In one condition, the subject was asked to recognize the vowel. In the other condition, the subject was asked to identify the speaker. Extensive training was provided for the speaker recognition task. Normal-hearing subjects achieved nearly perfect performance in both tasks. Cochlear-implant subjects achieved good performance in vowel recognition but poor performance in speaker recognition. The level of the cochlear implant performance was functionally equivalent to normal performance with eight spectral bands for vowel recognition but only to one band for speaker recognition. These results show a disassociation between speech and speaker recognition with primarily temporal cues, highlighting the limitation of current speech processing strategies in cochlear implants. Several methods, including explicit encoding of fundamental frequency and frequency modulation, are proposed to improve speaker recognition for current cochlear implant users.     

Voice conversion using structured Gaussian mixture model in cepstrum eigenspace

A new methodology of voice conversion in cepstrum eigenspace based on structured Gaussian mixture model is proposed for non-parallel corpora without joint training.For each speaker,the cepstrum features of speech are extracted,and mapped to the eigenspace which is formed by eigenvectors of its scatter matrix,thereby the Structured Gaussian Mixture Model in the EigenSpace(SGMM-ES)is trained.The source and target speaker's SGMM-ES are matched based on Acoustic Universal Structure(AUS)principle to achieve spectrum transform function.Experimental results show the speaker identification rate of conversion speech achieves95.25%,and the value of average cepstrum distortion is 1.25 which is 0.8%and 7.3%higher than the performance of SGMM method respectively.ABX and MOS evaluations indicate the conversion performance is quite close to the traditional method under the parallel corpora condition.The results show the eigenspace based structured Gaussian mixture model for voice conversion under the non-parallel corpora is effective.     

Acoustic effects of variation in vocal effort by men, women, and children

The acoustic effects of the adjustment in vocal effort that is required when the distance between speaker and addressee is varied over a large range (0.3-187.5 m) were investigated in phonated and, at shorter distances, also in whispered speech. Several characteristics were studied in the same sentence produced by men, women, and 7-year-old boys and girls: duration of vowels and consonants, pausing and occurrence of creaky voice, mean and range of F0, certain formant frequencies (F1 in [a] and F3), sound-pressure level (SPL) of voiced segments and [s], and spectral emphasis. In addition to levels and emphasis, vowel duration, F0, and F1 were substantially affected. "Vocal effort" was defined as the communication distance estimated by a group of listeners for each utterance. Most of the observed effects correlated better with this measure than with the actual distance, since some additional factors affected the speakers' choice. Differences between speaker groups emerged in segment durations, pausing behavior, and in the extent to which the SPL of [s] was affected. The whispered versions are compared with the phonated versions produced by the same speakers at the same distance. Several effects of whispering are found to be similar to those of increasing vocal effort.