1 code implementation • ACL 2022 • Junjie Chen, Xiangheng He, Yusuke Miyao
In this paper, we propose a mixture model-based end-to-end method to model the syntactic-semantic dependency correlation in Semantic Role Labeling (SRL).
no code implementations • 18 Apr 2024 • Junjie Chen, Xiangheng He, Danushka Bollegala, Yusuke Miyao
In this study, we empirically verify that \textbf{constituents correspond to word sequence patterns in the PAS-equivalent sentence set}.
no code implementations • 18 Sep 2023 • Xiangheng He, Junjie Chen, Björn W. Schuller
Our proposed method is significantly superior in terms of UAR and F1 to the single-task and multi-task baselines with p-values < 0. 05.
no code implementations • 6 Oct 2022 • Andreas Triantafyllopoulos, Björn W. Schuller, Gökçe İymen, Metin Sezgin, Xiangheng He, Zijiang Yang, Panagiotis Tzirakis, Shuo Liu, Silvan Mertes, Elisabeth André, Ruibo Fu, JianHua Tao
Speech is the fundamental mode of human communication, and its synthesis has long been a core priority in human-computer interaction research.
no code implementations • 10 May 2022 • Xiangheng He, Andreas Triantafyllopoulos, Alexander Kathan, Manuel Milling, Tianhao Yan, Srividya Tirunellai Rajamani, Ludwig Küster, Mathias Harrer, Elena Heber, Inga Grossmann, David D. Ebert, Björn W. Schuller
Previous studies have shown the correlation between sensor data collected from mobile phones and human depression states.
no code implementations • 6 May 2022 • Alexander Kathan, Andreas Triantafyllopoulos, Xiangheng He, Manuel Milling, Tianhao Yan, Srividya Tirunellai Rajamani, Ludwig Küster, Mathias Harrer, Elena Heber, Inga Grossmann, David D. Ebert, Björn W. Schuller
Digital health applications are becoming increasingly important for assessing and monitoring the wellbeing of people suffering from mental health conditions like depression.
1 code implementation • 18 Jul 2021 • Xiangheng He, Junjie Chen, Georgios Rizos, Björn W. Schuller
Emotional Voice Conversion (EVC) aims to convert the emotional style of a source speech signal to a target style while preserving its content and speaker identity information.