Objective: Cardiac auscultation is the most practiced non-invasive and cost-effective procedure for the early diagnosis of heart diseases. While machine learning based systems can aid in automatically screening patients, the robustness of these systems is affected by numerous factors including the stethoscope/sensor, environment and data collection protocol. This paper studies the adverse effect of domain variability on heart sound classification and develops strategies to address this problem. Methods: We propose a novel Convolutional Neural Network (CNN) layer, consisting of time-convolutional (tConv) units, that emulate Finite Impulse Response (FIR) filters. The filter coefficients can be updated via backpropagation and be stacked in the front-end of the network as a learnable filterbank. These filters can incorporate properties such as linear/zero phase-response and symmetry while improving robustness due to domain variability. Results: Our methods are evaluated using multi-domain heart sound recordings obtained from publicly available phonocardiogram (PCG) datasets. On multi-domain evaluation tasks, the proposed method surpasses the top-scoring systems found in the literature for heart sound classification. Our systems achieved relative improvements of up to 11.84% in terms of a modified accuracy (Macc) metric, compared to state-of-theart methods. Conclusion: The results demonstrate the effectiveness of the proposed learnable filterbank CNN architecture in achieving robustness towards sensor/domain variability in PCG signals. Significance: The proposed methods pave the way for deploying automated cardiac screening systems in diversified and underserved communities.

No abstract

We investigate two deep learning architectures reported to have superior performance in ASR over the conventional GMM system, with respect to automatic speech scoring. We use an approximately 800-hour large-vocabulary non-native spontaneous English corpus to build three ASR systems. One system is in GMM, and two are in deep learning architectures -namely, DNN and Tandem with bottleneck features. The evaluation results show that the both deep learning systems significantly outperform the GMM ASR. These ASR systems are used as the front-end in building an automated speech scoring system. To examine the effectiveness of the deep learning ASR systems for automated scoring, another non-native spontaneous speech corpus is used to train and evaluate the scoring models. Using deep learning architectures, ASR accuracies drop significantly on the scoring corpus, whereas the performance of the scoring systems get closer to human raters, and consistently better than the GMM one. Compared to the DNN ASR, the Tandem performs slightly better on the scoring speech while it is a little less accurate on the ASR evaluation dataset. Furthermore, given the results of the improved scoring performance while using fewer scoring features, the Tandem system shows more robustness for scoring task than the DNN one.Index Termsautomatic speech scoring, non-native spontaneous speech, automatic speech recognition, deep neural network, bottleneck features

Recent advancements in audio event classification often ignore the structure and relation between the label classes available as prior information. This structure can be defined by ontology and augmented in the classifier as a form of domain knowledge. To capture such dependencies between the labels, we propose an ontology-aware neural network containing two components: feed-forward ontology layers and graph convolutional networks (GCN). The feed-forward ontology layers capture the intra-dependencies of labels between different levels of ontology. On the other hand, GCN mainly models interdependency structure of labels within an ontology level. The framework is evaluated on two benchmark datasets for singlelabel and multi-label audio event classification tasks. The results demonstrate the proposed solutions efficacy to capture and explore the ontology relations and improve the classification performance.

This study focuses on acoustic variations in speech introduced by whispering, and proposes several strategies to improve robustness of automatic speech recognition of whispered speech with neutral-trained acoustic models. In the analysis part, differences in neutral and whispered speech captured in the UT-Vocal Effort II corpus are studied in terms of energy, spectral slope, and formant center frequency and bandwidth distributions in silence, voiced, and unvoiced speech signal segments. In the part dedicated to speech recognition, several strategies involving front-end filter bank redistribution, cepstral dimensionality reduction, and lexicon expansion for alternative pronunciations are proposed. The proposed neutral-trained system employing redistributed filter bank and reduced features provides a 7.7 % absolute WER reduction over the baseline system trained on neutral speech, and a 1.3 % reduction over a baseline system with whisper-adapted acoustic models.