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

Self-assembled biomimetic nanocomposite integrating defect-enhanced piezoelectricity with NIR-II fluorescence and MR imaging for ultrasound-triggered piezo/chemodynamic therapy of subcutaneous glioma.

Journal of nanobiotechnology·2026
Same author

Task-dependent neural dynamics in hippocampal CA1 during auditory and visual perceptual decision-making in rats.

Brain structure & function·2026
Same author

An enhanced state convergence architecture incorporating disturbance observer for bilateral teleoperation systems.

International journal of advanced robotic systems·2026
Same author

Personalized temporal interference targeting right striatum enhances structure-function couplings in cortico-striato-thalamo-cortical network.

Cerebral cortex (New York, N.Y. : 1991)·2026
Same author

Personalized high-intensity temporal interference stimulation decouples cerebellar networks to enhance implicit learning.

Journal of neuroengineering and rehabilitation·2026
Same author

GloW-VSNet: A scribble-based weakly supervised framework for global-view vitiligo lesion segmentation.

Medical image analysis·2025

相关实验视频

Updated: Jul 4, 2025

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

1.5K

扬声器转向意识日记化用于基于语音的认知评估.

Sean Shensheng Xu1, Xiaoquan Ke2, Man-Wai Mak2

  • 1School of Biomedical Engineering, Shenzhen University Medical School, Shenzhen University, Shenzhen, China.

Frontiers in neuroscience
|January 31, 2024
PubMed
概括

这项研究增强了演讲者对认知评估的日记化. 这种新方法在具有挑战性的条件下提高了准确性,并且可以识别演讲者的转换,帮助分析语音数据.

关键词:
莫卡 (MOCA) 是一种可怕的植物.这是一个全面的评分.痴呆症检测 痴呆症检测 痴呆症检测演讲者日记化 演讲者日记化扬声器嵌入式嵌入式发言人转时间

更多相关视频

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages
06:04

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages

Published on: March 24, 2023

394
Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education
09:00

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Published on: August 16, 2024

771

相关实验视频

Last Updated: Jul 4, 2025

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

1.5K
Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages
06:04

Systematic Hearing Performance Evaluation Process for Adolescents with Cochlear Implantation at Early Ages

Published on: March 24, 2023

394
Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education
09:00

Author Spotlight: Validation of SICOLE-R for Assessing Cognitive and Reading Skills in Spanish-Speaking Children and Its Role in Personalized Education

Published on: August 16, 2024

771

科学领域:

  • 语音处理 语音处理
  • 认知科学是一种认知科学.
  • 生物医学工程 生物医学工程

背景情况:

  • 在蒙特利尔认知评估 (MoCA) 等认知评估中,语音日记对于分析语音至关重要.
  • 传统方法面临声学不匹配和准确识别MoCA录音中的扬声器转换的挑战.

研究的目的:

  • 提议和评估三种用于MoCA录音中的演讲者日记的改进方法.
  • 为了提高日记的准确性和针对各种不匹配条件的稳定性.
  • 为了使数据集缺乏此信息的扬声器转换时间假设.

主要方法:

  • 使用多尺度通道相互依赖的扬声器嵌入与Res2Net块,挤压和激发单元,以及依赖通道的注意力.
  • 应用序列比较方法来实现整体的对话视图,以创建一个扬声器转向意识的评分矩阵.
  • 整合一对相似度计,将本地和全球信息整合到分数矩阵中.

主要成果:

  • 提议的改进在扬声器日记化方面明显优于传统的x-vector/PLDA系统.
  • 在语言,年龄和麦克风不匹配的场景下表现出卓越的性能.
  • 成功假设了扬声器转时间,提高了对各种数据集的适用性.

结论:

  • 开发的扬声器日记化系统为语音认知障碍诊断提供了更高的准确性和稳定性.
  • 该方法假设演讲者转换的能力使其对分析语音数据集非常有价值,即使是没有明确时间信息的语音数据集.