20 Papers
5 Citations
Xingfeng Li is an academic researcher from Japan Advanced Institute of Science and Technology. The author has contributed to research in topics: Computer science & Cluster analysis. The author has an hindex of 4, co-authored 7 publications.
Chat about Author
Papers
Speech Emotion Recognition Using 3D Convolutions and Attention-Based Sliding Recurrent Networks With Auditory Front-Ends
TL;DR: The experiments indicate that the proposed joint deep learning model that combines 3D convolutions and attention-based sliding recurrent neural networks (ASRNNs) for emotion recognition can be effectively used to recognize the emotions of speech from temporal modulation cues.
Auto-weighted Tensor Schatten p-Norm for Robust Multi-view Graph Clustering
TL;DR: In this article , a tensor-singular value decomposition based tensor nuclear norm (T-TNN) was proposed to better approximate the target rank of the learned graph tensor and make full use of prior information of singular values.
52
Improving multilingual speech emotion recognition by combining acoustic features in a three-layer model
Xingfeng Li,Masato Akagi +1 more
TL;DR: This study uses a three-layer model composed of acoustic features, semantic primitives, and emotion dimensions to map acoustics into emotion dimensions and classify the continuous emotion dimensional values into basic categories by using the logistic model trees.
47
Cross-view Graph Matching Guided Anchor Alignment for Incomplete Multi-view Clustering
TL;DR: This paper proposes Cross-view Graph Matching guided Anchor Alignment (CGMAA) to address cross-view anchor misalignment in incomplete multi-view clustering, improving clustering performance by predefining an anchor graph and unifying CGMAA with bipartite graph tensor learning.
37
A Three-Layer Emotion Perception Model for Valence and Arousal-Based Detection from Multilingual Speech.
Xingfeng Li,Masato Akagi +1 more
- 02 Sep 2018
TL;DR: This study proposes a framework to design, implement, and validate an emotion detection system using multiple corpora that outperformed the existing state-of-the-art system by yielding a smaller mean absolute error and higher correlation between estimates and human evaluators.
12