可重现的基于机器学习的语音病理检测:引入音调差异特征
Jan Vrba1, Jakub Steinbach1, Tomáš Jirsa2
1Department of Mathematics, Informatics, and Cybernetics, University of Chemistry and Technology, Technická 5, Prague 166 28, Czech Republic; Department of Radiological Imaging and Informatics, Tohoku University Graduate School of Medicine, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan.
Journal of voice : official journal of the Voice Foundation
|April 12, 2025
概括
这项研究引入了一种使用机器学习和新型声学特征检测语音病理学的新方法. 该方法实现了高回忆率,证明了诊断语音障碍的潜力.
科学领域:
- 语音和听力科学 语言和听力科学
- 计算语言学 计算语言学
- 生物医学工程 生物医学工程
背景情况:
- 语音病理会影响沟通和生活质量.
- 准确检测语音障碍对于及时干预至关重要.
- 现有的方法可能缺乏稳定性或依赖于复杂的特征集.
研究的目的:
- 开发和验证一种用于检测语音病理学的新方法.
- 利用公开可用的语音数据库和全面的功能集.
- 探索各种机器学习算法的有效性.
主要方法:
- 使用萨尔布鲁肯语音数据库进行分析.
- 结合既定的声学特征与新的音调差异和NaN特征.
- 评估了六个机器学习算法 (SVM,k-NN,天真贝叶斯,决策树,随机森林,AdaBoost).
- 使用网格搜索进行超参数优化和广泛的特征子集选择.
- 应用了k-means合成少数群体过量抽样技术来解决阶级不平衡.
- 使用重复分层交叉验证验证的验证模型.
主要成果:
- 实现了高的未加权平均回忆率:女性为85.61%,男性为84.69%,合计为85.22%.
- 证明了拟议的功能工程和机器学习方法的有效性.
- 由于其在不平衡的数据集中的偏差而遗漏了准确度指标.
结论:
- 拟议的方法表明,使用机器学习来检测语音病理的巨大潜力.
- 这种方法即使在简单的声乐任务 (持续的 /a:/ 元音) 中也有效.
- 提供一个公开可用的 GitHub 存储库和 REFORMS 检查列表,以提高可重复性和可用性.
相关概念视频
01:21Perceiving Loudness, Pitch, and Location
164
The human brain perceives pitch through two primary mechanisms reflected in place theory and frequency theory. Each mechanism describes how sound waves are interpreted as specific pitches by the brain, offering insights into the intricate processes of auditory perception.
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
Place theory, or place coding, suggests that different pitches are heard because various sound waves activate specific locations along the cochlea's basilar membrane. The brain determines the pitch of a sound by...
164
01:05Difference from Background: Limit of Detection
4.9K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
The LOD indicates the presence or absence...
4.9K
01:31Respiratory System Abnormal Finding II: Palpation and Auscultation
211
In assessing respiratory abnormalities, palpation and auscultation are critical tools for detecting and interpreting various pathophysiological changes. These techniques provide insight into underlying disorders by evaluating tactile sensations and sounds produced by the respiratory system.
Palpation Findings
During a respiratory assessment, palpation can reveal several vital abnormalities:
Palpation Findings
During a respiratory assessment, palpation can reveal several vital abnormalities:
211
01:30Classification of Signals
321
In signal processing, signals are classified based on various characteristics: continuous-time versus discrete-time, periodic versus aperiodic, analog versus digital, and causal versus noncausal. Each category highlights distinct properties crucial for understanding and manipulating signals.
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
A continuous-time signal holds a value at every instant in time, representing information seamlessly. In contrast, a discrete-time signal holds values only at specific moments, often denoted as x(n), where...
321
01:28Physical Assessment of the Respiratory Tract IV: Auscultation
227
Auscultation is a crucial component of the physical assessment of the respiratory tract. It offers valuable insights into airflow through the bronchial tree and potential lung obstructions. This process involves careful listening to breath, voice, and adventitious sounds, which can reveal a wealth of information about a patient's respiratory health.
Breath Sounds
Breath sounds are categorized into vesicular, bronchovesicular, and bronchial.
Breath Sounds
Breath sounds are categorized into vesicular, bronchovesicular, and bronchial.
227
01:00Detection of Gross Error: The Q Test
4.7K
When one or more data points appear far from the rest of the data, there is a need to determine whether they are outliers and whether they should be eliminated from the data set to ensure an accurate representation of the measured value. In many cases, outliers arise from gross errors (or human errors) and do not accurately reflect the underlying phenomenon. In some cases, however, these apparent outliers reflect true phenomenological differences. In these cases, we can use statistical methods...
4.7K