Neural network based methods have obtained great progress on a variety of natural language processing tasks. However, it is still a challenge task to model long texts, such as sentences and documents. In this paper, we propose a multi-timescale long short-termmemory (MT-LSTM) neural network to model long texts. MTLSTM partitions the hidden states of the standard LSTM into several groups. Each g...

To achieve high quality synthetic emotional speech, unitselection is the state-of-the-art technique. Nevertheless, a large expensive phonetically-segmented corpus is needed, and cost-effective automatic techniques should be studied. According to the HMM experiments in this paper: segmentation performance can depend heavily on the segmental or prosodic nature of the intended emotion (segmental e...

BACKGROUND Prevalence of chronic hepatitis C virus (HCV) infection is high among incarcerated persons in the United States. New, short-duration, high-efficacy therapies may expand treatment eligibility in this population. OBJECTIVE To assess the cost-effectiveness of sofosbuvir for HCV treatment in incarcerated populations. DESIGN Markov model. DATA SOURCES Published literature and expert...

Acta Academica is an accredited, open access South African journal dedicated to scholarship in the humanities. The publishes independently refereed research articles It promotes perspectives of critical social theory and engagements with postcolonial post-developmental debates special references (Southern) Africa. thereby support scholarly work that examines how humanities twenty-first century ...

In the present study, the ERP (event-related brain potentials) technique was used to investigate how accentuation influences the semantic short-term memory representations during on-line speech processing, and how this accentuation effect interacts with the distance of accentuation in the speech signal. Chinese spoken sentences were used as stimuli. The sentences included two critical words: No...