-
Article
Open AccessStripe-Transformer: deep stripe feature learning for music source separation
Music source separation (MSS) is to isolate musical instrument signals from the given music mixture. Stripes widely exist in music spectrograms, which potentially indicate high-level music information. For exa...
-
Chapter and Conference Paper
SepMLP: An All-MLP Architecture for Music Source Separation
Most previous deep learning based methods use convolutional neural networks (CNNs) or Recurrent neural networks (RNNs) to model the separation process in music signals. In this paper, we propose a MLP-like enc...
-
Chapter and Conference Paper
Improving Pathological Voice Detection: A Weakly Supervised Learning Method
Deep learning methods are data-driven. But for pathological voice detection, it is difficult to obtain high-quality labeled data. In this work, a weakly supervised learning Method is presented to improve the q...
-
Chapter
Multi-level Inter-regional Migrant Population Estimation Using Multi-source Spatiotemporal Big Data: A Case Study of Migrants in Hubei Province During the Outbreak of COVID-19 in Wuhan
Preventing the spread of COVID-19 is the most urgent need for society now. Before the closure of Wuhan on January 23, 2020, more than 5 million people had left Wuhan to other regions. A better understanding of...
-
Article
Open AccessEPHA4 haploinsufficiency is responsible for the short stature of a patient with 2q35-q36.2 deletion and Waardenburg syndrome
Waardenburg syndrome type I (WS1), an auditory-pigmentary genetic disorder, is caused by heterozygous loss-of-function mutations in PAX3. Abnormal physical signs such as dystopia canthorum, patchy hypopigmentatio...