Despite several decades of research, automatic speech recognition (ASR) lacks the performance achieved by human listeners. One of the major challenges in ASR is to cope with the immense variability of spoken language, which can be categorized into extrinsic sources (e.g., additive noise) and intrinsic factors (such as speaking rate, style, effort, dialect, and accent). What can we learn from th...

The strength of the acoustic startle response (ASR) to short bursts of broadband noise or tone pips (4, 8 and 16 kHz) and the prepulse inhibition (PPI) of the ASR elicited by prepulse tones (4, 8 and 16 kHz) were measured in parvalbumin-deficient (PV-/-) mice and in age-matched PV+/+ mice as controls. Hearing thresholds as determined from recordings of auditory brainstem responses were found to...

BACKGROUND Female cancer, especially breast and gynecologic cancers are considered multistage disease, highly influenced by risk and protective factors and/or screening preventive modalities. Consequences of all these factors result in the trend of change over time. METHODS In this comparative study, based on data of national cancer registry of Iran 2004 published by Iranian Ministry of Healt...

In this paper, we propose a mathematical study to simulate the dynamics of alkali–silica reaction (ASR) by using Caputo fractional derivative. We solve non-linear fractional-order system containing six differential equations understand ASR. For proving existence unique solution, use some recent novel properties Mittag–Leffler function along with fixed point theory. The stability proposed is als...

BACKGROUND The intracranial aneurysm (IA) size has been proved to have impacts on the hemodynamics and can be applied for the prediction of IA rupture risk. Although the relationship between aspect ratio and hemodynamic parameters was investigated using real patients and virtual models, few studies focused on longitudinal experiments of IAs based on patient-specific aneurysm models. We attempte...

Automatic Speech Recognition (ASR) systems can be trained to achieve remarkable performance given large amounts of manually transcribed speech, but labeled data sets difficult or expensive acquire for all languages interest. In this paper, we review the research literature identify models and ideas that could lead fully unsupervised ASR, including sub-word word modeling, segmentation speech sig...