Výsledky vyhledávání - "Liang, Mangui"

Report

Soft-Weighted CrossEntropy Loss for Continous Alzheimer's Disease Detection

Autor: Zhang, Xiaohui, Fu, Wenjie, Liang, Mangui

Alzheimer's disease is a common cognitive disorder in the elderly. Early and accurate diagnosis of Alzheimer's disease (AD) has a major impact on the progress of research on dementia. At present, researchers have used machine learning methods to dete

Externí odkaz: http://arxiv.org/abs/2402.11931

Zobrazit plný text záznamu

Report

Multimodal Emotion Recognition from Raw Audio with Sinc-convolution

Autor: Zhang, Xiaohui, Fu, Wenjie, Liang, Mangui

Speech Emotion Recognition (SER) is still a complex task for computers with average recall rates usually about 70% on the most realistic datasets. Most SER systems use hand-crafted features extracted from audio signal such as energy, zero crossing ra

Externí odkaz: http://arxiv.org/abs/2402.11954

Zobrazit plný text záznamu

Report

TST: Time-Sparse Transducer for Automatic Speech Recognition

Autor: Zhang, Xiaohui, Liang, Mangui, Tian, Zhengkun, Yi, Jiangyan, Tao, Jianhua

Publikováno v: International Conference on Artificial Intelligence (CICAI 2023)

End-to-end model, especially Recurrent Neural Network Transducer (RNN-T), has achieved great success in speech recognition. However, transducer requires a great memory footprint and computing time when processing a long decoding sequence. To solve th

Externí odkaz: http://arxiv.org/abs/2307.08323

Zobrazit plný text záznamu