英国 Laryngograph 电子声门仪 EGG-D100

太原理工大学

地区：中国

署名作者：朱婷

研究方向：情感病理语音数据库、视频、功能性近红外光谱成像、运动性构音障碍

关联论文 2 篇

英国纽卡斯尔大学

地区：英国

署名作者：梁慧智

研究方向：情感语音数据库、多模态情感识别、维度情感空间、三维电磁发音仪

关联论文 1 篇

中国太原市中心医院

地区：中国

署名作者：张伟

研究方向：情感病理语音数据库、视频、功能性近红外光谱成像、运动性构音障碍

关联论文 1 篇

构音障碍患者情感病理语音多模态同步采集与情感识别实验

方法说明：该实验使用 Laryngograph EGG-D100 电子声门仪同步采集脑卒中后运动性构音障碍患者与健康对照者在快乐、悲伤、愤怒、中性四种情绪下的声学信号和声门信号，同时配合 SONY FDR 摄像机、22 通道 fNIRS 与指尖脉搏血氧仪采集视频、脑血氧和生理信号，构建首个中文多模态情感病理语音数据库。随后利用微信小程序进行可懂度、情感类别与维度情感标注，并基于 IS09 特征集与 SVM/RF 分类器在声学与声门数据上开展自动情感识别与病情严重程度识别。

实验仪器：电子声门仪 EGG-D100；数码摄像机 FDR 系列；功能性近红外光谱成像系统；指尖脉搏血氧仪

实验步骤

1. 受试者招募与情绪诱发材料准备

   说明：在太原招募 39 名脑卒中后构音障碍患者和 29 名典型对照者（40～80 岁，普通话使用者），选取 27 句情感句和 9 句中性句，采用视听结合的情感片段与斯坦尼斯拉夫斯基情绪记忆法诱发快乐、悲伤、愤怒、中性四种情绪。

   关键点：患者与对照组年龄区间匹配；情绪诱发材料需兼顾情感强度与句子可读性

2. EGG-D100 电极佩戴与多设备布置

   说明：受试者佩戴 EGG-D100 电子声门仪，坐于距显示屏约 3 m 处，SONY FDR 摄像机置于约 50 cm 处，同时佩戴 22 通道 fNIRS 系统与指尖脉搏血氧仪；采集在受试者住所内进行并保证低噪声环境。

   关键点：电极需紧贴甲状软骨两侧以保证声门信号质量；环境低噪声；多设备同步启动

3. 声学与声门信号同步录制

   说明：通过 EGG-D100 同步获取声学信号和声门信号，受试者按提示依次朗读情感句和中性句，同时录制视频、fNIRS 与心率/血氧数据。

   关键点：EGG 左声道为声门信号、右声道为声学信号，二者严格同步

4. 数据完整性检查与整理

   说明：使用 Cool Edit Pro 检查声学与声门录音的完整性和可懂度，剔除无效样本，最终保留 4350 条声学数据和 4304 条声门数据。

   关键点：逐条核验录音完整性；声门数据与声学数据一一对应

5. 主观标注

   说明：通过自主开发的微信小程序对全部情感语音进行可懂度、离散情感类别及效价-唤醒度维度情感标注，并采集 SCL-90 量表。

   关键点：多标注者一致性；维度标注与离散标注同时进行

6. 特征提取与自动识别

   说明：提取 IS09 特征集（384 维）用于情感识别，提取包含 Pitch、PSD、LSF、MFCC（65 维）、LPC 统计量的 101 维特征用于病情严重程度识别，采用 SVM 与随机森林分类器分别在声学与声门数据上训练测试。

   关键点：声学与声门数据分别建模比较；对照组与患者组结果对比

实验数据

方法关联论文：情感多模态病理语音数据库的设计、构建与评估

普通话多模态情感语音数据库（发音运动学、声学、声门与面部微表情）构建与情感识别实验

方法说明：该实验使用 Laryngograph EGG-D100 电子声门仪与 EMA-AG501 三维电磁发音仪、SONY FDR-AX100 摄像机同步采集 22 名大学生在中性、高兴、愉快、冷漠、愤怒、悲伤、悲痛七种情绪下的声门信号、声学信号、发音运动学数据和面部微表情视频，建立 STEM-E2VA 普通话多模态情感语音数据库。信号同时标注离散情感标签和愉悦度-唤醒度-优势度维度标签，并基于 INTERSPEECH 2009 情感挑战特征集与 SVM、DNN、CNN 分类器分别评估声学、声门和运动学数据的情感识别性能。

实验仪器：电子声门仪 EGG-D100；三维电磁发音仪 AG501；数码摄像机 FDR-AX100

实验步骤

1. 语料设计与受试者选择

   说明：选取 24 个韵母和 16 个约 5 秒的陈述句作为语料，招募 22 名 21～27 岁、男女比例 1:1、母语为普通话的非表演专业大学生。

   关键点：语料需覆盖典型韵母与自然陈述句；受试者无表演训练背景以保证情感自然

2. 情绪诱发

   说明：结合自传体回忆情绪诱发法与想象情绪诱发法，引导受试者进入中性、高兴、愉快、冷漠、愤怒、悲伤、悲痛七种情绪状态。

   关键点：七种情绪在愉悦度-唤醒度-优势度空间中需有可区分的分布

3. EGG-D100 电极佩戴与校准

   说明：实验前由录制人员先行佩戴 EGG-D100，确保电极紧贴声门位置且不影响受试者正常发音；EGG 采集信号分为左声道（声门）和右声道（声学），左声道信号与声带接触面积正相关。

   关键点：电极位置需贴近甲状软骨两侧；佩戴不得妨碍发音

4. 多模态同步录制

   说明：使用 EMA-AG501 以 250 Hz 采集发音运动学数据、以 45 kHz 采集音频，SONY FDR-AX100 以 3840×2160、25 fps 录制面部视频，EGG-D100 同步采集声门与声学信号。

   关键点：各模态数据时间同步；音频最终降采样至 48 kHz、16 bit

5. 数据处理与标注

   说明：对录制数据进行分割整理，并标注离散情感标签（七类）与维度情感标签（愉悦度、唤醒度、优势度），验证维度标注数据的有效性。

   关键点：离散与维度标签一致性校验

6. 特征提取与情感识别

   说明：对声学与声门数据提取 INTERSPEECH 2009 情感挑战特征集（384 维），对运动学数据提取 28 维发音运动学特征，按 80%/20% 划分训练/测试集，采用 SVM、DNN、CNN 进行七类情感识别。

   关键点：分别评估单模态识别性能；比较不同分类器结果

实验数据

方法关联论文：普通话多模态情感语音数据库的构建与评估

情感多模态病理语音数据库的设计、构建与评估

原文题名：Design, construction and evaluation of emotional multimodal pathological speech database

发表年份：2023

作者：朱婷；段淑斐；梁慧智；张伟

刊源：arXiv

Larynograph

简介：I'm David Howard from york and we're company at the Dance center in London to look at B-boxing Let's look briefly at what the human instrument offers to the B-boxer because the human vocal instrument is the same for all of us We've all got what is basically a buzzer sounding something like this in our throat This is the vibrating vocal folds in the larynx so this can substitute for the larynx.

发布时间：2020-02-19T07:15:59Z

VOICED-UNVOICED CLASSIFICATION OF SPEECH USING AUTOCORRELATION MATRIX

简介：In this paper, a fast method for voiced-unvoiced classification of speech signals is introduced. The suggested method makes the V-UV decision, using signal energy, the peak-to-peak differerence of the autocorrelation function, number of zero crossings of the autocorrelation function and the unit delay autocorrelation coefficient all together.

发布时间：2020-02-11T07:12:57Z

Voice Timing in Two Parkinson’s Disease (PD) Patients following Subthalamic Nucleus Deep Brain Stimulation

简介：Speech and voice can have a variable response (Tripoliti et al, 2014). Subtypes of presentation have been proposed (Tsuboi et al, 2014). This study sought to examine the effects of STN-DBS on the temporal prosodic structure of voicing in the connected speech of two STN-DBS patients with contrasting presentations.

发布时间：2020-02-11T06:48:20Z

The Effect of Hearing Loss on the Vocal Features of Children

简介：The current study aimed to evaluate the effect of hearing loss on acoustic properties of voice through acoustic analysis of some vocal parameters in two groups of children with moderately severe or deep hearing loss, and normal children of the same age and gender.

发布时间：2020-02-11T06:22:23Z

REAL-TIME SPEECH PATTERN ELEMENT DISPLAYS FOR INTERACTIVE THERAPY

简介：New developments are presented and discussed for the interactive display and separate measurement of frication and nasality in addition to voice quality. Different clinical problems require appropriate combinations of speech pattern elements support different approaches and these are supported by appropriate combinations of speech pattern elements. Particular use has been made of the laryngograph signal so that the displays can be given a sense of immediacy and are highly accurate.

发布时间：2020-02-11T05:52:34Z

Pitch Determination from Bone Conducted Speech

简介：This paper explores the potential of pitch determination from bone conducted (BC) speech. Pitch determination from normal air conducted (AC) speech signal can not attain the expected level of accuracy for every voice and background conditions. In contrast, since BC speech is caused by the vibrations that have traveled through the vocal tract wall, it is robust against ambient conditions.

发布时间：2020-02-11T04:14:42Z

NON-INVASIVE MONITORING OF REFLEXIVE SWALLOWING

简介：These methods do not use radiation and have the potential to give more accurate timing information than can be derived from auditory/acoustic monitoring. Pilot data were obtained from the simultaneous use of four sensors: an ear-plug microphone of the type used successfully for the detection of otoacoustic emissions; a standard miniature electret microphone ordinarily used for speech recording; a miniature accelerometer of the type sometimes used for monitoring nasality....

发布时间：2020-02-11T03:40:53Z

Measuring voice in the clinic-Laryngograph Speech Studio Analyses

简介：In the present work the application of pitch perceptual criteria is described in the provision of an integrated framework for practical clinical assessment and therapy. Sustained sounds are shown to require quite different levels of sampling accuracy from those needed for ordinary connected speech. Radically useful results come from the application of these basic ideas to the representation and analysis of the pathological and the normal voice in case management and audit.

发布时间：2020-02-11T03:05:31Z

Laryngograph as a tool in diagnosis and therapy of voice disorders

简介：The electroglottography method is an instrumental analysis by the device called Laryngograph, which can be practically applied in the speech signal analysis for people with voice disorders issues. In this paper capabilities of this device are estimated. A very interesting fact is that visualization of the speech signal obtained with the use of Laryngograph device allows to detect the acoustically and phonetically most important features and present them in a graphical form.

发布时间：2020-02-11T02:07:14Z

SPEECHREADING OF WORDS AND SENTENCES BY NORMALLY HEARING AND HEARING IMPAIRED CHINESE SUBJECTS: THE ENHANCEMENT EFFECTS OF COMPOUND SPEECH PATTERNS

简介：Speech perception requires a receiver to make decisions both about trends and also about the language patterns of a speaker. Hearing people use the speaker’s speech signal as the direct sensory evidence. In the special case of visual speech perception, otherwise known as speechreading, this sensory evidence is derived from the visible articulation movements of speech. Unfortunately, many important articulation movements are invisible.

发布时间：2020-02-10T10:41:36Z

Closing and Opening Phase Variability in Dysphonia

简介：Four examples of the use of vocal fold contact phase measurement are discussed for unilateral paresis. In each case this aspect of voice quality is of greater importance than the physical measurement of loudness and pitch related parameters. For three of the cases electrostimulation has been used as a main part of the treatment.

发布时间：2020-02-10T09:48:51Z

Automatic Evaluation of Voice Quality Using Text-Based Laryngograph Measurements and Prosodic Analysis

简介：Automatic Evaluation of Voice Quality Using Text-Based Laryngograph Measurements and Prosodic Analysis

发布时间：2020-02-10T08:56:30Z

ASPECTS OF VOICE QUALITY: DISPLAY, MEASUREMENT AND THERAPY

简介：In this workshop the physical measures of voice fundamental frequency and larynx contact quotient are explored in relation to the perception, analysis and treatment of aspects of voice quality in a speech and language therapy clinic. A fundamental principle is to analyse samples of continuous speech. The workshop will focus on clients with unilateral vocal fold paralysis.

发布时间：2020-02-10T07:13:37Z

Acoustical characteristics of vocal modes in singing

简介：This paper looks further into the characteristics of the voice modes in singing in order to test the model already provided. The conclusion is that the model is too simple to cover the full range. The work has also provided information on singers’ SPL and formants’ repositioning in dependence of pitch. Further work is recommended.

发布时间：2020-02-10T06:35:19Z

A Longitudinal Study Investigating Synchronization in a Singing Quintet

简介：Research suggests that synchronization between musicians during ensemble performances can be affected by the rhythmic or tonal complexity of the piece being performed and by group roles such as leader-follower relationships. Since previous studies have mostly been conducted within single performance sessions, developmental aspects of interpersonal synchronization in ensembles remain underinvestigated.

发布时间：2020-02-10T04:45:14Z

A corpus of read and conversational Austrian German

简介：This paper presents GRASS (Graz corpus of Read And Spontaneous Speech), the first large scale speech database for Austrian German with both read and conversational speech. In total, the corpus contains approximately 1900 minutes of speech in which 38 speakers produced more than 220 000 word tokens from 14 593 di↵erent word types. The corpus consists of three components.

发布时间：2020-02-10T03:22:46Z