Mohan Xu — accepted AI-conference papers

2025

TIGER: Time-frequency Interleaved Gain Extraction and Reconstruction for Efficient Speech Separation

ICLR 2025poster

In recent years, much speech separation research has focused primarily on improving model performance. However, for low-latency speech processing systems, high efficiency is equally important. Therefore, we propose a speech separation model with significantly reduced parameters and computational cos…

Cited by 0 Source ↗PDF ↗Scholar ↗

2023

Hierarchical Transformer for Multi-Label Trailer Genre Classification

Zihui Cai, Hongwei Ding, Xuemeng Wu, Mohan Xu, Xiaohui Cui

ICASSP 2023accepted

Determining the genres of a trailer is a challenging multi-label classification task. Previous studies tend to classify by CNN or RNN. Recently, Transformer based on attention mechanism has achieved better results in many research fields than CNN and RNN. Inspired by these, we propose a Hierarchical…

Cited by 0 Source ↗Scholar ↗