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

Cognitive load weakens neural speech tracking without altering response timing.

Communications biology·2026
Same author

Charge-Reversal Nano-Drug Delivery Systems in the Tumor Microenvironment: Mechanisms, Challenges, and Therapeutic Applications.

International journal of molecular sciences·2024
Same author

A Low-Noise-Level Heart Sound System Based on Novel Thorax-Integration Head Design and Wavelet Denoising Algorithm.

Micromachines·2019
Same author

[Arthroscopic reconstruction of anterior cruciate ligament with preservation of the remnant bundle].

Zhongguo gu shang = China journal of orthopaedics and traumatology·2013
Same author

[Anterior cruciate ligament reconstruction with tendon graft enveloped by preserved remnants].

Zhongguo gu shang = China journal of orthopaedics and traumatology·2013
Same author

Genetic and molecular biological characterization of two homologous cheR genes from Leptospira interrogans.

Acta biochimica et biophysica Sinica·2013

相关实验视频

Updated: Jun 10, 2025

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants
11:14

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

10.9K

ADT网络:一种新的非线性方法,用于从EEG信号中解码语音封面.

Ruixiang Liu1, Chang Liu1, Dan Cui2

  • 1School of Intelligent Medicine, China Medical University, Shenyang, China.

Trends in hearing
|October 14, 2024
PubMed
概括

我们开发了一种听觉解码变压器 (ADT) 网络,以精确地从电脑电图 (EEG) 信号中解码语音信封. 这种可解释的深度学习模型显示了客观的听觉处理评估和听力损失诊断的前景.

关键词:
深度学习是一种深度学习.一个电脑电图 (electroencephalogram) 是一个电脑电图.可以解释的解释性.神经信封跟踪跟踪神经信封跟踪变压器变压器变压器变压器

更多相关视频

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

43.3K
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
08:51

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

5.6K

相关实验视频

Last Updated: Jun 10, 2025

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants
11:14

A Novel Experimental and Analytical Approach to the Multimodal Neural Decoding of Intent During Social Interaction in Freely-behaving Human Infants

Published on: October 4, 2015

10.9K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

43.3K
Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms
08:51

Statistical Modelling of Cortical Connectivity Using Non-invasive Electroencephalograms

Published on: November 1, 2019

5.6K

科学领域:

  • 神经科学是一个神经科学.
  • 信号处理 信号处理
  • 机器学习 机器学习

背景情况:

  • 对听力处理的客观评估对于诊断听力障碍至关重要.
  • 目前用于从脑电图 (EEG) 信号中解码语音信封的方法缺乏高精度和可解释性.
  • 开发先进的计算模型可以增强我们对神经对语音反应的理解.

研究的目的:

  • 提出一种新的深度学习模型,即听觉解码变压器 (ADT) 网络,用于从EEG信号中准确和可解释的语音外重建.
  • 根据现有的非线性模型,评估ADT网络的性能.
  • 通过ADT网络的解释性特征,分析语音信封跟踪的基础神经机制.

主要方法:

  • 使用深度学习方法,结合时空卷积来提取特征,并使用变压器解码器来解码语音信封.
  • 实施反因果掩盖,以确保模型仅考虑当前和未来的EEG特征,模仿自然的语音-EEG关系.
  • 视觉化时空卷积权重作为时间域过器和大脑拓,以及对时空卷积内核的废除研究.

主要成果:

  • 获得了0.168 (SparrKULee) 和0.167 (DTU) 的竞争平均重建得分,与其他非线性模型相比.
  • 确定低 (0.5-8 Hz) 和高频 (14-32 Hz) EEG信号对于语音外重建至关重要.
  • 证明活跃的大脑区域主要是双边的,位于听觉皮层内,与先前的研究一致.
  • 注意分数可视化进一步证实了听觉处理研究中的现有发现.

结论:

  • 听觉解码变压器 (ADT) 网络提供了一种平衡的方法来实现高性能和可解释的语音封面解码来自EEG.
  • 该模型的可解释性为神经语音信封跟踪机制提供了洞察力.
  • ADT网络为客观的听觉处理评估和听力损失诊断的潜在进展提供了一个有希望的工具.