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

Introduction to Cognitive Psychology01:20

Introduction to Cognitive Psychology

Cognitive psychology is the field of psychology dedicated to examining how people think. It attempts to explain how and why we think the way we do by studying the interactions among human thinking, emotion, creativity, language, and problem-solving, as well as other cognitive processes. Cognitive psychology studies how information is processed and manipulated in remembering, thinking, and knowing.
This field emerged in the mid-20th century, following a period dominated by behaviorism, which...

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

The protective mechanism of sevoflurane in pulmonary arterial hypertension via downregulation of TRAF6.

Toxicology and applied pharmacology·2024
Same author

The comparison of gut microbiota between wild and captive Asian badgers (Meles leucurus) under different seasons.

Scientific reports·2024
Same author

Plasma Proteomics Analysis of Early Biomarkers for Predicting Female Fecundability: A Nested Case-Control Study.

Journal of proteome research·2024
Same author

Serum 5'-Nucleotidase as a Novel Predictor of Adverse Clinical Outcomes after Percutaneous Coronary Intervention in Patients with Coronary Artery Disease.

Reviews in cardiovascular medicine·2024
Same author

Evidence of a distinct collective mode in Kagome superconductors.

Nature communications·2024
Same author

Effect of lumbosacral transitional vertebra on developmental alterations of the hip: a quantitative investigation of the lumbo-pelvic-hip complex via whole-body computed tomography.

Quantitative imaging in medicine and surgery·2024

相关实验视频

Updated: May 13, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.7K

可解释的AI用于认知和计算神经科学中的法医语音认证.

Zhe Cheng1, Haitao Yang1,2, Yingzhuo Xiong1

  • 1Department of Criminal Investigation, Hunan Police Academy, Changsha, China.

Frontiers in neuroscience
|November 21, 2025
PubMed
概括

本研究介绍了一种使用卷积神经网络 (CNN) 和长短期记忆 (LSTM) 网络的深度学习框架,以进行强大的深度虚假音频检测. 可解释的AI方法证实了模型.

关键词:
真正的检测检测是真实的认知神经科学 认知神经科学数字语音处理器是数字语音处理器.可解释的人工智能多媒体法医学

更多相关视频

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
09:09

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

785
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

相关实验视频

Last Updated: May 13, 2026

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.7K
Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
09:09

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

785
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

科学领域:

  • 数字法医学数字法医学
  • 人工智能的人工智能
  • 语音处理 语音处理

背景情况:

  • 深度假冒音频对法医语音认证构成重大威胁.
  • 现有的检测方法与操纵音频的复杂性作斗争.

研究的目的:

  • 开发一个深度学习框架,用于更好地检测 deepfake 音频.
  • 提高操纵音频检测系统的准确性和概括性.

主要方法:

  • 一种混合深度学习模型,将卷积神经网络 (CNN) 结合起来用于光谱特征提取和长短期记忆 (LSTM) 网络用于时间建模.
  • 作为声学特征使用的线性频率 Cepstral 系数 (LFCC).
  • 应用可解释的人工智能 (XAI) 技术,如Grad-CAM和SHAP,以实现模型可解释性.

主要成果:

  • 在ASVspoof2019 LA和WaveFake数据集上,CNN-LSTM模型实现了卓越的准确性和概括性.
  • 在LFCC的特点中,其性能优于Mel-Frequency Cepstral Coefficients (MFCC) 和Gammatone Frequency Cepstral Coefficients (GFCC) 的表现.
  • XAI分析显示,该模型的重点是高频器件和时间不一致,验证了其法医相关性.

结论:

  • 拟议的框架为法医音频认证提供了一个强大而可解释的解决方案.
  • 集成XAI提高了深度假冒检测的信任和透明度.
  • LFCC 功能在识别操纵的音频方面具有显著的法医价值.