Short-segment heart sound classification using an ensemble of deep convolutional neural networks [PDF]
, 2018 This paper proposes a framework based on deep convolutional neural networks (CNNs) for automatic heart sound classification using short-segments of individual heart beats. We design a 1D-CNN that directly learns features from raw heart-sound signals, and
Noman, Fuad +3 more
core +2 more sources
Sound Event Detection with Sequentially Labelled Data Based on Connectionist Temporal Classification and Unsupervised Clustering [PDF]
, 2019 Sound event detection (SED) methods typically rely on either strongly labelled data or weakly labelled data. As an alternative, sequentially labelled data (SLD) was proposed.
Hou, Yuanbo +3 more
core +3 more sources
Acoustic Scene Classification by Implicitly Identifying Distinct Sound Events
, 2019 In this paper, we propose a new strategy for acoustic scene classification (ASC) , namely recognizing acoustic scenes through identifying distinct sound events.
Deng, Shiwen +3 more
core +1 more source
Vestibular Patient Journey: Insights From Vestibular Disorders Association (VeDA) Registry
Annals of Clinical and Translational Neurology, EarlyView.ABSTRACT Objective Vestibular symptoms impose a high burden of disability. Understanding real‐world diagnostic and treatment pathways can identify care gaps and guide interventions. We aimed to characterize symptom profiles, diagnostic trends, provider involvement, and treatment patterns in vestibular disorders.
Ali Rafati +10 more
wiley +1 more source
Audio Classification from Time-Frequency Texture
, 2008 Time-frequency representations of audio signals often resemble texture images. This paper derives a simple audio classification algorithm based on treating sound spectrograms as texture images.
Slotine, Jean-Jacques, Yu, Guoshen
core +1 more source
Elevated Connectivity During Language Processing Is Associated With Cognitive Performance in SeLECTS
Annals of Clinical and Translational Neurology, EarlyView.ABSTRACT Objective Self‐Limited Epilepsy with Centrotemporal Spikes (SeLECTS) is associated with language impairments despite seizures originating in the motor cortex, suggesting aberrant cross‐network interactions. Here we tested whether functional connectivity in SeLECTS during language tasks predicts language performance.
Wendy Qi +8 more
wiley +1 more source
Continuous robust sound event classification using time-frequency features and deep learning.
PLoS ONE, 2017 The automatic detection and recognition of sound events by computers is a requirement for a number of emerging sensing and human computer interaction technologies.
Ian McLoughlin +5 more
doaj +1 more source
Animal Sound Classification Using Dissimilarity Spaces
Applied Sciences, 2020 The classifier system proposed in this work combines the dissimilarity spaces produced by a set of Siamese neural networks (SNNs) designed using four different backbones with different clustering techniques for training SVMs for automated animal audio ...
Loris Nanni +3 more
doaj +1 more source
Visualization and categorization of ecological acoustic events based on discriminant features
Ecological Indicators, 2021 Although sound classification in soundscape studies are generally performed by experts, the large growth of acoustic data presents a major challenge for performing such task.
Liz Maribel Huancapaza Hilasaca +3 more
doaj +1 more source
Improving acoustic vehicle classification by information fusion
, 2011 We present an information fusion approach for ground vehicle classification based on the emitted acoustic signal. Many acoustic factors can contribute to the classification accuracy of working ground vehicles.
Damarla, Thyagaraju +2 more
core +1 more source