Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关实验视频

Updated: Jan 11, 2026

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

2.0K

使用自主监督变压器和多任务学习的扬声器独立性脱节性关节症严重程度分类.

Balasundaram Kadirvelu1, Lauren Stumpf1, Sigourney Waibel1

  • 1Brain & Behaviour Lab, Department of Computing and Department of Bioengineering, Imperial College London, London, United Kingdom.

PLOS digital health
|November 12, 2025
PubMed
概括
此摘要是机器生成的。

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Digital Phenotyping for Adolescent Mental Health: Feasibility Study Using Machine Learning to Predict Mental Health Risk From Active and Passive Smartphone Data.

Journal of medical Internet research·2026
Same author

Safety of human-AI cooperative decision-making within intensive care: A physical simulation study.

PLOS digital health·2025
Same author

The 'Sandwich' meta-framework for architecture agnostic deep privacy-preserving transfer learning for non-invasive brainwave decoding.

Journal of neural engineering·2024
Same author

Eye tracking insights into physician behaviour with safe and unsafe explainable AI recommendations.

NPJ digital medicine·2024
Same author

Quantifying the impact of AI recommendations with explanations on prescription decision making.

NPJ digital medicine·2023
Same author

EEG decoding for datasets with heterogenous electrode configurations using transfer learning graph neural networks.

Journal of neural engineering·2023
Same journal

Vision-language models for human motion understanding: Lessons from stroke rehabilitation.

PLOS digital health·2026
Same journal

A digital marker for stratifying cardiovascular metabolic comorbidities among the middle-aged and elderly adults.

PLOS digital health·2026
Same journal

User engagement in the tuberculosis treatment support tools intervention and its impact on treatment outcomes: A secondary analysis of a pragmatic trial.

PLOS digital health·2026
Same journal

Machine learning for risk stratification of hypertensive disorders of pregnancy: Enhancing clinical efficiency in low-resource antenatal care in Tanzania.

PLOS digital health·2026
Same journal

The trust in AI-generated health advice (TAIGHA) scale and short version (TAIGHA-S): Development and validation study.

PLOS digital health·2026
Same journal

Time-series prediction of adverse birth outcomes in the U.S. using multilayer perceptron neural networks.

PLOS digital health·2026
查看所有相关文章

一个新的机器学习框架,语音不可知潜伏规范化 (SALR),提供了对神经疾病中常见的言语障碍 - - 脱节症的客观评估. SALR提高了语音严重性分类的准确性,为传统方法提供了具有成本效益的替代方案.

科学领域:

  • 神经语言学和计算听力学
  • 医疗保健中的人工智能
  • 语音病理学和康复治疗 语音病理学和康复治疗

背景情况:

  • 发音障碍是由于神经疾病导致的语言障碍,由于其复杂性和主观评估方法,它提出了诊断挑战.
  • 目前的临床评估依赖于专家视听分析,这可能是耗时和不一致的.
  • 需要客观的定量方法来准确地分层和监测失关节症的严重程度.

研究的目的:

  • 引入一种新的机器学习框架,即扬声器不可知潜伏规范化 (SALR),用于对失联症的客观分类和监测.
  • 开发一个独立于说话者的模型,减少对个人的言语特征的依赖.
  • 提供一种可访问和具有成本效益的工具,用于评估失联症的严重程度.

主要方法:

  • 使用基于变压器的架构,集成在健康语音数据上预先训练的 wav2vec 2.0 模型.
  • 实施了具有多任务目标的对比性学习策略:严重程度分类的交叉损失和语音不可知嵌入的三倍边缘损失.
  • 采用扬声器不可知潜伏规范化 (SALR) 来按严重程度分组潜伏语音嵌入,而不是按扬声器.

主要成果:

  • 在UA-Speech数据集上,SALR框架实现了70.5%的准确性和59.2%的F1得分,使用了离开一个主体的交叉验证.
  • 与之前的基准指标相比,显著改善了16.5%的绝对值 (30%的相对值).

更多相关视频

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.3K

相关实验视频

Last Updated: Jan 11, 2026

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

2.0K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

2.3K
  • 可解释性分析证实了隐性空间中增强的顺序结构,减少了扬声器特定的依赖性,并显示了强度.

    • 结论:

    • 萨尔框架显示了独立于说话者的失联症严重程度分类的巨大潜力.
    • 这种方法提供了一个客观的,可访问的,和成本效益的替代传统的肌痛评估方法.
    • 该SALR框架对语音障碍评估中的自动化临床应用具有有前途的意义.