强大的声质特征嵌入用于失声语音检测
Jianwei Zhang1, Julie Liss2, Suren Jayasuriya3
1School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85281, USA.
概括
这项研究引入了一种新的深度学习框架,用于准确自动检测语音障碍 (失声症). 该方法产生了强大的声学嵌入,改善了不同数据集和条件的性能.
科学领域:
- 语音处理 语音处理
- 生物医学工程 生物医学工程
- 机器学习 机器学习
背景情况:
- 大约1.2%的全球人口经历了语音产生受损,需要可靠的自动化评估工具.
- 当前的自动语音分析方法往往缺乏跨不同数据集和应用程序的概括性.
- 对于失声语音检测有很大需求的强大而准确的方法.
研究的目的:
- 开发一个深度学习框架,用于生成对声质敏感的声学特征嵌入.
- 增强语音分析模型的稳定性,跨越不同的身体和条件.
- 为了提高自动失声语音检测的准确性和通用性.
主要方法:
- 通过对比和分类损失函数的组合来训练一个深度学习模型.
- 数据曲技术应用于输入语音样本,以增加模型的稳定性.
- 该框架旨在生成对语音质量敏感的声学特征嵌入.
主要成果:
- 拟议的方法实现了高分类准确性,无论是在不同公司内部还是跨越不同公司.
- 生成的嵌入显示了对不同数据集的语音质量和稳定性的敏感性.
- 该模型在干净和恶化的语音数据集上始终超过了三个基线方法.
结论:
- 开发的深度学习框架为自动失声语音检测提供了强大而准确的解决方案.
- 该方法在体内通用的能力使其适用于各种临床和研究应用.
- 产生的声学嵌入有潜力进一步推进语音质量评估.
更多相关视频
06:24Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications
Published on: January 5, 2024
875
09:09Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
Published on: September 27, 2024
464
相关概念视频
Perceiving Loudness, Pitch, and Location
220
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...
220
Pulse amplitude and quality
1.8K
Pulse amplitude is a crucial indicator of cardiac health because it provides valuable insights into the strength of left ventricular contractions and the overall uniformity of blood circulation within the vasculature. The strength of the pulse is directly related to the force with which the heart contracts and the volume of blood being pumped.
A weak or absent pulse may indicate reduced cardiac output or poor left ventricular contraction, which can be signs of cardiovascular dysfunction or...
A weak or absent pulse may indicate reduced cardiac output or poor left ventricular contraction, which can be signs of cardiovascular dysfunction or...
1.8K
Force Classification
1.2K
Forces play a crucial role in the study of physics and engineering. They are essential in describing the motion, behavior, and equilibrium of objects in the physical world. Forces can be classified based on their origin, type, and direction of action.
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
1.2K
Unsoundness of Aggregate due to Volume Change
115
Unsoundness in aggregates due to volume changes is primarily caused by the physical alterations aggregates undergo, such as freezing and thawing, thermal changes, and wetting and drying. Unsound aggregates, when subjected to these changes, result in volume change upon disintegration. This, in turn, contributes to the deterioration of concrete, including scaling, pop-outs, and cracking. Particular types of aggregates, such as porous flints, cherts, and those containing clay minerals, are...
115
Downsampling
167
When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
167
Classification of Signals
484
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...
484
