In this study, we assessed the reduction in the computational costs of a room-acoustics solver by the partition of unity finite-element method (PUFEM), particularly addressing the element matrix construction process with numerical integration rules. The PUFEM enriches the approximation of sound fields by incorporating a general solution of the Helmholtz equation into shape functions. Plane wave enrichment is applied herein. In plane-wave-enriched FEM, the construction of element matrices using a high-order Gauss–Legendre quadrature is the main numerical operation with a long computational time. To reduce the computational time of the room-acoustics solver with plane-wave-enriched FEM, in this report, we describe our exploration of efficient integration rules via an ideal plane wave propagation problem in a duct. We present two integration rules: a well-used existing rule extended to the low-frequency range and another derived by the linear regression of the relationship between the number of wavelengths included in each element and the minimum number of integration points required for solution convergence. Numerical results revealed that both rules produce accurate frequency responses in a broad frequency range. However, the rule obtained by linear regression outperforms the extended rule.

Osteoarthritis (OA) of the knee is a widespread disease caused by the articular cartilage damage, and its prevalence has become a severe public health problem worldwide, especially in the ageing society. Although X-ray, MRI, CT, etc. are commonly used to examine knee OA by inserting external high energy into the body, they do not provide dynamic information on knee joint integrity. In the present research, the acoustic emission (AE) technique has been applied in healthy individuals as well as OA patients in order to evaluate the knee integrity in dynamic analysis modes without inserting any external energy. Four groups of people, young, middle-aged, older, and OA patient have been participated in the present research, and significant results have been identified. It has been found that the degeneration of the articular cartilage progresses gradually with the increase of the age. The angular positions of knee damage are also evaluated by clarifying AE hits. The results are verified through clinical investigations by an orthopedic surgeon applying X-Ray and MRI techniques. The results of the present research demonstrate that the AE technique can be considered as a promising tool for the diagnosis of knee osteoarthritis.

Research in corpus-driven Automatic Speech Recognition (ASR) is advancing rapidly towards building a robust Large Vocabulary Continuous Speech Recognition (LVCSR) system. Under-resourced languages like Bangla require benchmarking large corpora for more research on LVCSR to tackle their limitations and avoid the biased results. In this paper, a publicly published large-scale Bangladeshi Bangla speech corpus is used to implement deep Convolutional Neural Network (CNN) based model and Recurrent Neural Network (RNN) based model with Connectionist Temporal Classification (CTC) loss function for Bangla LVCSR. In experimental evaluations, we find that CNN-based architecture yields superior results over the RNN-based approach. This study also emphasizes assessing the quality of an open-source large-scale Bangladeshi Bangla speech corpus and investigating the effect of the various high-order N-gram Language Models (LM) on a morphologically rich language Bangla. We achieve 36.12% word error rate (WER) using CNN-based acoustic model and 13.93% WER using beam search decoding with 5-gram LM. The findings demonstrate by far the state-of-the-art performance of any Bangla LVCSR system on a specific benchmarked large corpus.

Phase recovery is a methodology of estimating a phase spectrogram that is reasonable for a given amplitude spectrogram. For enhancing the signals obtained from the processed amplitude spectrograms, it has been applied to several audio applications such as harmonic/percussive source separation (HPSS). Because HPSS is often utilized as preprocessing of other processes, its phase recovery should be simple. Therefore, practically effective methods without requiring much computational cost, such as phase unwrapping (PU), have been considered in HPSS. However, PU often results in a phase that is completely different from the true phase because (1) it does not consider the observed phase and (2) estimation error is accumulated with time. To circumvent this problem, we propose a phase-recovery method for HPSS using the observed phase information. Instead of accumulating the phase as in PU, we formulate a local optimization model based on the observed phase so that the estimated phase remains similar to the observed phase. The analytic solution to the proposed optimization model is provided to keep the computational cost cheap. In addition, iterative refinement of phase in the existing methods is applied for further improving the result. From the experiments, it was confirmed that the proposed method outperformed PU.

This paper presents a proposal for an efficient room acoustic solver with dissipation-free and dispersion-optimized explicit time-domain FEM (TD-FEM), with investigation of its applicability for broadband room acoustic modeling from three numerical experiments. Recently, FEM-based room acoustic solvers have attracted great attention because of their strength in handling complex geometries. However, the development of higher-efficiency solvers is unavoidable to perform acoustic modeling of real-size rooms at kilohertz frequency with small discretization error: the dispersion error. The present paper first formulates a novel room acoustic solver with dissipation-free fourth-order accurate explicit TD-FEM using a three-step time integration method. A dispersion-optimized solver is further proposed in which dispersion error is minimized in the axial and diagonal directions at a specific frequency under given spatial resolution mesh or elements, by which the approximation capability at higher frequencies is enhanced without any additional computational cost. The performance of the optimized solver in broadband acoustic simulation using cubic elements is then examined in comparison with the original fourth-order accurate solver and the standard implicit TD-FEM. Finally, higher efficiency of the optimized solver is also demonstrated for acoustic simulation in a larger rectangular room discretized with rectangular and distorted hexahedral elements.