Jianhua Tao

ORCID: 0000-0002-0477-587X
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Coastal and Marine Dynamics
  • Ocean Waves and Remote Sensing
  • Music and Audio Processing
  • Water Quality Monitoring and Analysis
  • Oceanographic and Atmospheric Processes
  • Water Quality Monitoring Technologies
  • Marine and coastal ecosystems
  • Coastal wetland ecosystem dynamics
  • Tropical and Extratropical Cyclones Research
  • Fluid Dynamics Simulations and Interactions
  • Coastal and Marine Management
  • Wave and Wind Energy Systems
  • Water Quality and Pollution Assessment
  • Digital Media Forensic Detection
  • Speech Recognition and Synthesis
  • Hydrological Forecasting Using AI
  • Environmental Quality and Pollution
  • Earthquake and Tsunami Effects
  • Emotion and Mood Recognition
  • Computational Fluid Dynamics and Aerodynamics
  • Environmental Changes in China
  • Generative Adversarial Networks and Image Synthesis
  • Speech and Audio Processing
  • Aquatic and Environmental Studies
  • Industrial Technology and Control Systems

Tsinghua University
2001-2025

Institute of Automation
2023

Anhui University
2023

Tianjin University
2010-2020

Institute of Remote Sensing and Digital Earth
2014

Chinese Academy of Sciences
2014

State Key Laboratory of Remote Sensing Science
2014

Beijing Center for Disease Prevention and Control
2013

Guangzhou University
2010

Ministry of Education of the People's Republic of China
2001

The first Multimodal Emotion Recognition Challenge (MER 2023)1 was successfully held at ACM Multimedia. challenge focuses on system robustness and consists of three distinct tracks: (1) MER-MULTI, where participants are required to recognize both discrete dimensional emotions; (2) MER-NOISE, in which noise is added test videos for modality evaluation; (3) MER-SEMI, provides a large amount unlabeled samples semi-supervised learning. In this paper, we introduce the motivation behind challenge,...

10.1145/3581783.3612836 article EN cc-by 2023-10-26

With the proliferation of Audio Language Model (ALM) based deepfake audio, there is an urgent need for generalized detection methods. ALM-based audio currently exhibits widespread, high deception, and type versatility, posing a significant challenge to current (ADD) models trained solely on vocoded data. To effectively detect we focus mechanism generation method, conversion from neural codec waveform. We initially constructed Codecfake dataset, open-source, large-scale collection comprising...

10.1109/taslpro.2025.3525966 article EN IEEE Transactions on Audio Speech and Language Processing 2025-01-01

10.1109/tkde.2025.3550545 article EN IEEE Transactions on Knowledge and Data Engineering 2025-01-01

10.1016/s1001-6058(06)60027-9 article EN Journal of Hydrodynamics 2006-07-01

This paper reports on an investigation of the spatial and temporal variation main environmental indicators in Pearl River estuary using a numerical water quality model. An integrated field survey was undertaken 2006 2007, data obtained were used to set up verify The model driven by hydrodynamic which monthly average freshwater discharges from eight major entrances taken into account seaward boundary elevations specified based tidal constituents. Two – dissolved inorganic nitrogen chemical...

10.1680/wama.12.00011 article EN Proceedings of the Institution of Civil Engineers - Water Management 2013-06-20

Although current fake audio detection approaches have achieved remarkable success on specific datasets, they often fail when evaluated with datasets from different distributions. Previous studies typically address distribution shift by focusing using extra data or applying loss restrictions during training. However, these methods either require a substantial amount of complicate the training process. In this work, we propose stable learning-based scheme that involves Sample Weight Learning...

10.48550/arxiv.2406.03237 preprint EN arXiv (Cornell University) 2024-06-05

Fake artefacts for discriminating between bonafide and fake audio can exist in both short- long-range segments. Therefore, combining local global feature information effectively discriminate audio. This paper proposes an end-to-end bidirectional state space model, named RawBMamba, to capture discriminative deepfake detection. Specifically, we use sinc Layer multiple convolutional layers short-range features, then design a Mamba address Mamba's unidirectional modelling problem further...

10.21437/interspeech.2024-698 article EN Interspeech 2022 2024-09-01

Abstract Bohai Bay is a large, semi-enclosed shallow water basin located along the western region of Sea in north-eastern part China. As result rapid economic development surrounding coastal hinterland, quality characteristics within have been growing concern recent years. Details are given herein an aquatic environmental assessment study undertaken to investigate hydrodynamic and transport processes using integrated numerical modelling approach tield measurements. 3-D model was used...

10.1080/00221686.2001.9628290 article EN Journal of Hydraulic Research 2001-12-01

A considerable body of experimental evidence has accumulated that suggests the circulations generated in numerical models free-surface Hows are two-dimensional plan may be realistic physically even though none ofthe experiments described literature have explicitely utilized lateral shear stresses. On other hand, it is easily shown by analytical means not only impossible to generate without introduction correspondingly stresses but that, absence such stresses, no circulation at all should...

10.1080/00221688509499342 article EN Journal of Hydraulic Research 1985-08-01

Abstract A numerical model of pollutant transport acted by water waves on a shallow‐water mild‐slope beach is established in this study. The combined with wave propagation model, multiple wave‐breaking wave‐induced current and convection–dispersion model. based the higher‐order approximation parabolic equation which can be used to simulate refraction, diffraction breaking large area near‐shore zone using concept radiation stress considering effect bottom resistance caused waves. verified...

10.1002/fld.1116 article EN International Journal for Numerical Methods in Fluids 2005-12-20
Coming Soon ...