正常、大声两种说话状态下语音的声学差异及对声纹鉴定的影响

本文主要通过对正常、大声两种说话状态下的普通话中三个单元音[a]、[i]、[u]的声强、时长、基频、谐波振幅差值、共振峰等声学参数的分析,综合比较了各参数的变化规律,发现大声说话时的语音并非正常语音的简单放大,二者不仅在声强上存在差别,同时在频率域上也发生了重要变化。同一人不同状态下发音的频谱特征差异性较大,同种状态下发音的相似性、可比性较强,为此,声纹鉴定中应尽量选取状态相同的语音进行比对。     

噪声环境下的说话人识别研究

论文研究了噪声环境下的基于GMM模型和MFCC参数的说话人识别方法.白噪声环境下,说话人识别的正确率大大降低.论文研究了GMM模型在说话人识别中的应用,并采用MFCC参数作为语音的特征参数进行噪声环境下的识别研究.实验结果表明,在0dB到15dB的噪声环境下,使用GMM模型进行说话人识别时,采用MFCC参数比采用LPCC特征参数能够获得更高的识别率.在10dB的低信噪比环境下,使用MFCC参数的说话人识别,仍然可以获得75.6％的识别率,可以有效进行识别.     

不同通信系统下的说话人识别探究

目的研究不同通信系统条件下,利用语音进行说话人识别的基本假设＂个体之间的语音差异大于个体自身的语音变异＂是否成立。方法选择宽带语图差异较大的不同通信系统下语音进行比对及差异的统计。结果发现说话人识别的基本假设在不同通信系统条件下成立。结论得出了不同通信系统条件下进行说话人识别的方法与判阈。     

汉语听力损伤儿童语言康复中语音教学的有关问题

一、问题的提出在听力损伤儿童(即人们习惯上所称的聋儿或聋童)语言康复训练中,语音教学具有重要意义,但迄今为止,怎样对汉语听力损伤儿童进行语音教学的问题尚未得到令人满意的解决。目前,对汉语听力损伤儿童的语音教学仍沿用国外的方法,例如,语言康复训练中基本上采用音素训练法,即教聋童从音素学起,教他们发元音[a]、[o]、[r]、[i]、     

伪装语音的自动话者识别研究

本文以语音自动话者识别系统为平台,对实际案件中出现频率较高的10种伪装语音类型进行自动话者识别测试。通过对20位发音人的正常语音和10种类型伪装语音的话者辨认和话者确认测试,分析不同语音伪装类型对自动话者识别的影响。该结果对于深入认识伪装语音的特性及其话者识别研究具有重要意义,也为自动话者识别技术的改进提供了参考。     

南京话响音的鼻化度——兼论/n、l/不分的实质

本文使用鼻音计(Nasometer)考察了南京话响音的鼻化度。结果发现:南京话元音内在鼻化度大小与元音发音时舌位的前、后和高、低有关系,表现出与北京话大体一致的趋势,都是[a]的鼻化度最高,[u]的鼻化度最低,前高元音[y]的鼻化度也较低。元音的内在鼻化度及非鼻音浊声母[l]、[Z]的鼻化度在40以上,鼻音声母[m]的鼻化度在80以上,鼻化元音基本都处在N值40和80之间的断裂带上。南京话鼻音[n]与边音[l]相混的现象实质是/n/与/l/两个音位的完全合流,[n]与[l]是一个音位的两个条件变体,合流过程中产生了中介音——鼻化边音[I]。     

法庭语音学

本文对法庭语音学进行了研究综述,主要介绍了该学科的核心内容:说话人鉴定。在实际办案中,当未找到嫌疑人,只有犯罪分子的检材语音时,可以使用说话人画像/说话人分类技术。若没有犯罪分子的录音证据时,可以让受害人和证人进行说话人的听觉辨认。具体的辨认形式有两种:对熟人辨认和对陌生人辨认,在对陌生人辨认时可以采用语音辨认的方法进行。当检材语音和样本语音都齐备的时候,法庭语音分析专家就可以对二者进行比对检验了。目前语音比对分析涉及到的问题和领域有:基于贝叶斯方法的法庭推理和似然比计算、共振峰频率的测量应用、非解析感知与样例理论、法庭说话人自动识别以及不同方法的综合应用等。     

法庭语音学

本文对法庭语音学进行了研究综述,主要介绍了该学科的核心内容：说话人鉴定。在实际办案中,当未找到嫌疑人,只有犯罪分子的检材语音时,可以使用说话人画像/说话人分类技术。若没有犯罪分子的录音证据时,可以让受害人和证人进行说话人的听觉辨认。具体的辨认形式有两种：对熟人辨认和对陌生人辨认,在对陌生人辨认时可以采用语音辨认的方法进行。当检材语音和样本语音都齐备的时候,法庭语音分析专家就可以对二者进行比对检验了。目前语音比对分析涉及到的问题和领域有：基于贝叶斯方法的法庭推理和似然比计算、共振峰频率的测量应用、非解析感知与样例理论、法庭说话人自动识别以及不同方法的综合应用等。     

言语习惯特征在声纹鉴定中的作用

言语习惯特征是人们在学习言语过程中,受环境因素和个体因素影响,逐渐形成的一类语音声学特征,它包括:方言、口音、惯用语、俚语、赘语,有的人说普通话时夹杂的地方方言特征;以及言语的速率、清晰度、流畅度、语音喉化、气嗓音、言语缺欠等[1]。大量实验和研究表明,对个人语音特征的形成起决定作用的,是发声系统的解剖学形状和发音方法的动态个人特点。从生理、心理学机制来看,言语的形成包括言语神经机制、言语的脑机制、言语的感知、思维和反映模式,是一个复杂的心理生理过程。每个人从“咿呀学话”开始,经过千百次“听想说”的学习、校练…     

手机通话语音的实验研究

当前,手机通话语音已成为司法语音鉴定中最为常见的一种语音形式。本研究从手机通信系统的信道特点出发,分析手机通话语音的声谱特点和共振峰频率变化等情况;同时还比较了不同通话网络、不同通话方式及不同手机的通话语音特点。实验发现,手机通话语音与直接录音语音有明显的变化,主要表现在高低频信息的带宽滤波效应、高低频共振峰的漂移、语音质量、音色、韵律特征等方面;还发现,不同手机通话条件下的语音变化程度不同。最后,讨论了手机通话语音变化对说话人鉴定的影响及鉴定中的注意事项。     

电子伪装语音的变声规律研究

本文以目前社会上流行的两款不同类别的电子伪装语音器材为对象,对多名实验对象变声前、后的声学特征（基频、共振峰、声调、能量和过零率等）进行了深入的实验分析。实验结果和分析表明,变声相对于原声,其声学特征变化是有规律的,根据这一声学特征变化规律,对变声语音逆变声就能得到与原声符合很好的恢复语音,这为深入开展对经过电子伪装的语音进行同一人鉴定奠定了基础。     

电子伪装语音的变声规律研究

本文以目前社会上流行的两款不同类别的电子伪装语音器材为对象,对多名实验对象变声前、后的声学特征(基频、共振峰、声调、能量和过零率等)进行了深入的实验分析。实验结果和分析表明,变声相对于原声,其声学特征变化是有规律的,根据这一声学特征变化规律,对变声语音逆变声就能得到与原声符合很好的恢复语音,这为深入开展对经过电子伪装的语音进行同一人鉴定奠定了基础。     

Voice stress evaluators and lie detection

Proponents of vocal stress analysis systems argue that they are able to detect spoken deception by analysis of "stress" in the voice signal. Presumably, they do so by examining traces made by laryngeal microtremors which, they claim exist in the voice, are associated with stress, and ultimately are associated with lying. However, most research that seeks to identify the relationships between microtremors and laryngeal function has produced negative results, and data on the ability of voice analyzers to detect stress from speech--or to identify spoken deception--have been negative or "mixed" in nature. Since perspectives based on available results leave a number of questions unanswered, a series of experiments has been undertaken. The first was focused on the basic acoustic/temporal correlates of stress in voice (the subject of an earlier report), the second on examination of stress by commercial voice analyzers, and the third on the detection of relatively high-risk lies by this same type of voice analysis procedure. It was found that correct stress/nonstress identifications occurred only at chance levels; the lie/nonlie identification scores were quite similar with professional "examiners" performing at about the same level of accuracy as other auditors. The following review is divided into two parts: a history of the controversy and a presentation of the two cited experiments.     

声纹鉴定中的音强特征研究

目的 确定声纹鉴定中的音强特征的应用价值.方法 将10名条件相近、普通话较标准的男性发音人的语音样本输入VS-99型语音工作站,分别对每个人所发的相同音节韵母的宽带声纹图谱进行分析比对.结果 不同人发出的相同音节,其音强分布一般不同,有时还会出现明显的差异,主要表现在韵母总体音强按频率分布状态、韵母总体音强按时间分布状...     

Voice disguise and automatic speaker recognition

In this paper a newly developed Forensic Automatic Speaker Recognition System (FASRS) was introduced and the effect of 10 types of voice disguises that are common in forensic casework on the performance of this system was studied. In this study 10 types of disguised voices and normal voices from 20 male college students were used as test samples. Each disguised voice was compared with all normal voices in the database to make speaker identification and speaker verification. The result of speaker recognition is summarized and the influence of voice disguises on the FASRS is evaluated.     

My Life for A Voice: The Influence of Voice on Health-Care Decisions

Are people willing to give up affordable healthcare and future years of their lives in exchange for having a voice in healthcare decision-making? Drawing upon research on the psychology of justice, we claim that the fairness of healthcare decision-making procedures, expressed by the availability of voice, can be more important than critical health-related outcomes. We examined this proposition using a forced-choice paradigm that required participants to choose between voice and better healthcare outcomes (affordable healthcare and greater life expectancy). Findings from three studies revealed that people maintain a strong preference for voice even at the expense of tangible healthcare outcomes. In study 1, participants preferred a healthcare plan that offered them a voice when it was $3,000–$12,000 more costly than a plan that did not offer such voice privileges. In study 2, participants preferred a voice plan to a no-voice plan when the no-voice plan was 5–20 years greater in its average life expectancy compared with the voice plan. In study 3, which used a more demographically diverse, non-student sample, the preference for the voice plan persisted across all conditions, even when the no-voice plan was 25 years greater in its life expectancy, and even when participants’ expected to personally live longer under the no-voice plan. These results are explained by participants’ expectation to enjoy better personal healthcare outcomes and greater autonomy when afforded voice. These findings demonstrate the importance of voice in hypothetical decision-making relevant to policy-making.     

Revisiting the Instrumentality of Voice: Having Voice in the Process Makes People Think They Will Get What They Want

Research on procedural justice has found that processes that allow people voice (i.e., input) are perceived as fairer, and thus elicit more positive reactions, than processes that do not allow people voice. Original theorizing attributed these effects to beliefs that the provision of voice enhances people’s sense of process control, which people were assumed to value because it impacts their perceived likelihood of receiving desired outcomes (the instrumental perspective of procedural justice). Subsequent research questioned this perspective, arguing that outcome expectations do not account for the effects of voice. However, this subsequent research failed to directly examine the interplay of voice, outcome expectations, and reactions. The current studies revisit and extend research on this topic by asking whether manipulations of voice act as shared circumstance effects. Confirming an untested implication of the instrumental perspective, we show that giving everyone voice increases their belief, ex-ante, that they are likely to win an upcoming competition. However, this instrumental belief accounts for only part of the effects of voice on perceived procedural fairness and on general reactions to outcomes. Results suggest that voice does indeed have instrumental significance, an implication not adequately recognized in current justice theorizing. However, this instrumentality does not, by itself, explain why people value having a voice in processes that affect them.