Search

Mode decomposition methods decompose a complex signal into several principal components, which are prone to be demodulated out of non-stationary features. The efficacy of current approaches deteriorates when the components spectrally overlap, owing to their limited frequency resolution and inadequate decomposition levels. To overcome these limitations, we propose a Singular Spectrum Mode Ensemble (SME) method as an extension of the Singular Spectrum Analysis (SSA), which was traditionally considered a linear approach. The SME method employs the high-rank Hankel matrix and Singular Value Decomposition (SVD), complemented by our proposed Fast Singular Spectrum Decomposition (FSSD) and a grouping method based on Singular Value-Frequency Band correlations. Theoretical research in this paper reveals that our method achieves exceptionally high resolution in the frequency domain, facilitating the separation of the components intermixed in the spectrum. Verification experiments demonstrate our method's ability to separate non-stationary components overlapping with others in the time-frequency domain, unveiling some components in mechanical and speech signals that were previously difficult to separate and discern.

Cross-domain speaker recognition (SR) can be improved by unsupervised domain adaptation (UDA) algorithms. UDA algorithms often reduce domain mismatch at the cost of decreasing the discrimination of speaker features. In contrast, optimal transport (OT) has the potential to achieve domain alignment while preserving the speaker discrimination capability in UDA applications; however, naively applying OT to measure global probability distribution discrepancies between the source and target domains may induce negative transports where samples belonging to different speakers are coupled in transportation. These negative transports reduce the SR model's discriminative power, degrading the SR performance. This paper proposes a coupling-regularized optimal transport (CROT) algorithm for cross-domain SR to reduce the negative transport during UDA. In the proposed CROT, two consecutive processing modules regularize the coupling paths for the OT solution: a progressive inter-speaker constraint (PISC) module and a coupling-smoothed regularization (CSR) module. The PISC, designed as a pseudo-label memory bank with curriculum learning, is first applied to select valid samples to guarantee that coupling samples are from the same speaker. The CSR, designed to control the information entropy of the coupling paths further, reduces the effect of negative transport in UDA. To evaluate the effectiveness of the proposed algorithm, cross-domain SR experiments were conducted under different target domains, speaker encoders, corpora, and acoustic features. Experimental results showed that CROT achieved a 50% relative reduction in equal error rates compared to conventional OT-based UDAs, outperforming the state-of-the-art UDAs.

The regulation of macrophage phenotype (M1/M2) is very important for tissue repair. The macrophage phenotypes could be affected by the physical and chemical parameters of implant surface. The aim of this study was to investigate the effects of surface modifications of titanium metals on macrophage phenotype. The medical pure titanium metals (PT-Ti) subjected to Anodic Oxidation (AO-Ti), Sand Blasting/acid etching (SLA-Ti) and Plasma-sprayed HA coating (HA coating-Ti) were used for regulating the phenotype of macrophage. The results showed that the Raw264.7 cells of AO-Ti groups had no obvious pseudopodia and could spread evenly in all directions. The levels of IL-1β and TNF-α, which belong to pro-inflammatory genes, expressed by the cells on AO-Ti groups were the lowest among all of the modified titanium groups. But, the levels of IL-10 and TGF-β, which belong to anti-inflammatory genes, expressed on AO-Ti groups were much higher than those on the other groups. Furthermore, the AO-Ti could regulate the expression of SOCS-1 and SOCS-3 to affect the active of NF-κB signaling. The gene expression results of macrophages showed that the AO-Ti was more conductive to inhibit the expression of M1-related genes and promote the expression of M2-related genes in an inflammatory environment. The AO-Ti was more beneficial to tissue repair than other modified titanium metals. The results showed that the anodic oxidation is an effective method to regulate the phenotype of macrophages.

Peri-implantitis are a major problem causing implant failure these days. Accordingly, anti-infection during the early stage and subsequent promotion of osseointegration are two main key factors to solve this issue. Micro-arc oxidation (MAO) treatment is a way to form an oxidation film on the surface of metallic materials. The method shows good osteogenic properties but weak antibacterial effect. Therefore, we developed combined strategies to combat severe peri-implantitis, which included the use of a novel compound, PD, comprising dendrimers poly(amidoamine) (PAMAM) loading dimethylaminododecyl methacrylate (DMADDM) as well as MAO treatment. Here, we explored the chemical properties of the novel compound PD, and proved that this compound was successfully synthesized, with the loading efficiency and encapsulation efficiency of 23.91% and 31.42%, respectively. We further report the two-stage double benefits capability of PD + MAO: (1) in the first stage, PD + MAO could decrease the adherence and development of biofilms by releasing DMADDM in the highly infected first stage after implant surgery both in vitroand in vivo; (2) in the second stage, PD + MAO indicated mighty anti-infection and osteoconductive characteristics in a rat model of peri-implantitis in vivo. This study first reports the two-staged, double benefits of PD + MAO, and demonstrates its potential in clinical applications for inhibiting peri-implantitis, especially in patients with severe infection risk.

The task of speaker verification (SV) is to decide whether an utterance is spoken by a target or an imposter speaker. In most studies of SV, a log-likelihood ratio (LLR) score is estimated based on a generative probability model on speaker features, and compared with a threshold for making a decision. However, the generative model usually focuses on individual feature distributions, does not have the discriminative feature selection ability, and is easy to be distracted by nuisance features. The SV, as a hypothesis test, could be formulated as a binary discrimination task where neural network based discriminative learning could be applied. In discriminative learning, the nuisance features could be removed with the help of label supervision. However, discriminative learning pays more attention to classification boundaries, and is prone to overfitting to a training set which may result in bad generalization on a test set. In this paper, we propose a hybrid learning framework, i.e., coupling a joint Bayesian (JB) generative model structure and parameters with a neural discriminative learning framework for SV. In the hybrid framework, a two-branch Siamese neural network is built with dense layers that are coupled with factorized affine transforms as used in the JB model. The LLR score estimation in the JB model is formulated according to the distance metric in the discriminative learning framework. By initializing the two-branch neural network with the generatively learned model parameters of the JB model, we further train the model parameters with the pairwise samples as a binary discrimination task. Moreover, a direct evaluation metric (DEM) in SV based on minimum empirical Bayes risk (EBR) is designed and integrated as an objective function in the discriminative learning. We carried out SV experiments on Speakers in the wild (SITW) and Voxceleb. Experimental results showed that our proposed model improved the performance with a large margin compared with state of the art models for SV.

Deep learning-based models have greatly advanced the performance of speech enhancement (SE) systems. However, two problems remain unsolved, which are closely related to model generalizability to noisy conditions: (1) mismatched noisy condition during testing, i.e., the performance is generally sub-optimal when models are tested with unseen noise types that are not involved in the training data; (2) local focus on specific noisy conditions, i.e., models trained using multiple types of noises cannot optimally remove a specific noise type even though the noise type has been involved in the training data. These problems are common in real applications. In this article, we propose a novel denoising autoencoder with a multi-branched encoder (termed DAEME) model to deal with these two problems. In the DAEME model, two stages are involved: training and testing. In the training stage, we build multiple component models to form a multi-branched encoder based on a decision tree (DSDT). The DSDT is built based on prior knowledge of speech and noisy conditions (the speaker, environment, and signal factors are considered in this paper), where each component of the multi-branched encoder performs a particular mapping from noisy to clean speech along the branch in the DSDT. Finally, a decoder is trained on top of the multi-branched encoder. In the testing stage, noisy speech is first processed by each component model. The multiple outputs from these models are then integrated into the decoder to determine the final enhanced speech. Experimental results show that DAEME is superior to several baseline models in terms of objective evaluation metrics, automatic speech recognition results, and quality in subjective human listening tests.

Magnesium (Mg), is widely used for the bone repair in oral and orthopedic application due to excellent bioactivity, degradation and biocompatibility. However, the range of application is greatly limited because of the rapid degradation of Mg metal in the body. Surface modification is an effective method to enhance the corrosion resistance and reduce the degradation rate of Mg metal. In the present study, pure Mg metal (P-Mg) was subjected to alkali-heat treatment (AT-Mg) or anodic oxidation-heat treatment (AO-HT-Mg). Both AT-Mg and AO-HT-Mg had a layer of MgO on their surfaces after treatment. Then the effects of MgO coating on corrosion resistance, bioactivity, Mesenchymal Stem Cells’ (MSCs) proliferation, adhesion and osteogenic differentiation, and the bone repair capability of Mg metal were investigated. We found both AT-Mg and AO-HT-Mg had stronger corrosion resistance than P-Mg. MSCs on both AT-Mg and AO-HT-Mg had higher expression of proteins and genes of ALP, OCN, Col-I and Runx2 than those on P-Mg. They also showed better bone repair property than P-Mg in vivo. In general, MgO layer formed by anodic oxidation-heat treatment had better resistance and biocompatibility than that produced by alkali-heat treatment. This study indicated the MgO coating not only improved the corrosion resistance of Mg metal, but also promoted the osteogenic differentiation of MSCs and bone regeneration.

Speech enhancement model is used to map a noisy speech to a clean speech. In the training stage, an objective function is often adopted to optimize the model parameters. However, in the existing literature, there is an inconsistency between the model optimization criterion and the evaluation criterion for the enhanced speech. For example, in measuring speech intelligibility, most of the evaluation metric is based on a short-time objective intelligibility (STOI) measure, while the frame based mean square error (MSE) between estimated and clean speech is widely used in optimizing the model. Due to the inconsistency, there is no guarantee that the trained model can provide optimal performance in applications. In this study, we propose an end-to-end utterance-based speech enhancement framework using fully convolutional neural networks (FCN) to reduce the gap between the model optimization and the evaluation criterion. Because of the utterance-based optimization, temporal correlation information of long speech segments, or even at the entire utterance level, can be considered to directly optimize perception-based objective functions. As an example, we implemented the proposed FCN enhancement framework to optimize the STOI measure. Experimental results show that the STOI of a test speech processed by the proposed approach is better than conventional MSE-optimized speech due to the consistency between the training and the evaluation targets. Moreover, by integrating the STOI into model optimization, the intelligibility of human subjects and automatic speech recognition system on the enhanced speech is also substantially improved compared to those generated based on the minimum MSE criterion.

This paper presents a novel decoding framework for acoustic models (AMs) based on end-to-end neural networks (e.g., connectionist temporal classification). The end-to-end training of AMs has recently demonstrated high accuracy and efficiency in automatic speech recognition (ASR). When using the trained AM in decoding, although a language model (LM) is implicitly involved in such an end-to-end AM, it is still essential to integrate an external LM trained with a large text corpus to achieve the best results. While there is no theoretical justification, most of the studies suggest using a naive interpolation of the end-to-end AM score and the external LM score, empirically. In this paper, we propose a more theoretically sound decoding framework derived from a maximization of the posterior probability of a word sequence given an observation. As a consequence of the theory, the subword LM is newly introduced to seamlessly integrate the external LM score with the end-to-end AM score. Our proposed method can be achieved by a small modification of the conventional weighted finite-state transducer-based implementation, without having to heavily increase the graph size. We tested the proposed decoding framework on ASR experiments with the Corpus of the Wall Street Journal and the Corpus of Spontaneous Japanese. The results showed that the proposed framework achieved significant and consistent improvements over the conventional interpolation-based decoding framework.