arXiv subjects

Ying Shi

Publications and source records attributed to Ying Shi.

At least 37 records · Page 2Linked to original sources

arXiv2023

Spot keywords from very noisy and mixed speech

Most existing keyword spotting research focuses on conditions with slight or moderate noise. In this paper, we try to tackle a more challenging task: detecting keywords buried under strong interfering speech (10 times higher than the keyword in amplitude), and even worse, mixed with other keywords. We propose a novel Mix Training (MT) strategy that encourages the model to discover low-energy keywords from noisy and mixed speech. Experiments were conducted with a vanilla CNN and two EfficientNet (B0/B2) architectures. The results evaluated with the Google Speech Command dataset demonstrated that the proposed mix training approach is highly effective and outperforms standard data augmentation and mixup training.

cs.SD↗

arXiv2023

Coherent Optical Spin Hall Transport for Spin-optronics at Room Temperature

Spin or valley degrees of freedom in condensed matter have been proposed as efficient information carriers towards next generation spintronics. It is therefore crucial to develop effective strategies to generate and control spin or valley-locked spin currents, e.g., by exploiting the spin Hall or valley Hall effects. However, the scattering, and rapid dephasing of electrons pose major challenges to achieve macroscopic coherent spin currents and realistic spintronic or valleytronic devices, specifically at room temperature, where strong thermal fluctuations could further obscure the spin flow. Exciton polaritons in semiconductor microcavities being the quantum superposition of excitons and photons, are believed to be promising platforms for spin-dependent optoelectronic or, in short, spin-optronic devices. Long-range spin current flows of exciton polaritons may be controlled through the optical spin Hall effect. However, this effect could neither be unequivocally observed at room temperature nor be exploited for realistic polariton spintronic devices due to the presence of strong thermal fluctuations or large linear spin splittings. Here, we report the observation of room temperature optical spin Hall effect of exciton polaritons with the spin current flow over a distance as large as 60 um in a hybrid organic-inorganic FAPbBr3 perovskite microcavity. We show direct evidence of the long-range coherence at room temperature in the flow of exciton polaritons, and the spin current carried by them. By harnessing the long-range spin-Hall transport of exciton polaritons, we have demonstrated two novel room temperature polaritonic devices, namely the NOT gate and the spin-polarized beam splitter, advancing the frontier of room-temperature polaritonics in perovskite microcavities.

cond-mat.mes-hall↗

arXiv2022

Matrix-valued Cauchy bi-orthogonal polynomials and a novel noncommutative integrable lattice

Matrix-valued Cauchy bi-orthogonal polynomials were proposed in this paper, together with its quasideterminant expression. It is shown that the coefficients in four-term recurrence relation for matrix-valued Cauchy bi-orthogonal polynomials should satisfy a novel noncommutative integrable system, whose Lax pair is given by fractional differential operators with non-abelian variables.

math-ph↗

arXiv2022

Solutions and continuum limits to nonlocal discrete sine-Gordon equations: bilinearization reduction method

As with nonlocal continuous and semi-discrete integrable systems, the study of nonlocal discrete integrable systems is also of interest. In this paper, local and nonlocal reductions of a fully discrete negative order Ablowitz-Kaup-Newell-Segur equation are investigated. We give out the exact solutions in double Casoratian form to the reduced nonlocal discrete sine-Gordon equations by the bilinearization reduction method. Then, through the continuum limits, nonlocal semi-discrete sine-Gordon equations and their solutions are obtained. The dynamics of soliton solutions are analyzed and illustrated by asymptotic analysis. The research ideas and methods in this paper can be generalized to promote the studies on nonlocal discrete integrable systems.

nlin.SI↗

arXiv2022

Magnon squeezing enhanced entanglement in a cavity magnomechanical system

We investigate the generation of the entanglement in a cavity magnomechanical system, which consists of three modes: a magnon mode, a microwave cavity mode and a mechanical vibration mode, the couplings of the magnon-photon and the magnon-phonon are achieved by the magnetic dipole interaction and the magnetostrictive interaction, respectively. By introducing a squeezing of the magnon mode, the magnon-photon and the magnon-phonon entanglements are significantly enhanced compared with the case without inserting the magnon squeezing. We find that an optimal parameter of the squeezing exists, which yields the maximum entanglement. This study provides a new idea for exploring the properties of quantum entanglement in the the cavity magnomechanical systems, and may have some potential applications in the quantum state engineering.

quant-ph↗

arXiv2020

Can We Trust Deep Speech Prior?

Recently, speech enhancement (SE) based on deep speech prior has attracted much attention, such as the variational auto-encoder with non-negative matrix factorization (VAE-NMF) architecture. Compared to conventional approaches that represent clean speech by shallow models such as Gaussians with a low-rank covariance, the new approach employs deep generative models to represent the clean speech, which often provides a better prior. Despite the clear advantage in theory, we argue that deep priors must be used with much caution, since the likelihood produced by a deep generative model does not always coincide with the speech quality. We designed a comprehensive study on this issue and demonstrated that based on deep speech priors, a reasonable SE performance can be achieved, but the results might be suboptimal. A careful analysis showed that this problem is deeply rooted in the disharmony between the flexibility of deep generative models and the nature of the maximum-likelihood (ML) training.

cs.SD↗

arXiv2019

Discriminative Embedding Autoencoder with a Regressor Feedback for Zero-Shot Learning

Zero-shot learning (ZSL) aims to recognize the novel object categories using the semantic representation of categories, and the key idea is to explore the knowledge of how the novel class is semantically related to the familiar classes. Some typical models are to learn the proper embedding between the image feature space and the semantic space, whilst it is important to learn discriminative features and comprise the coarse-to-fine image feature and semantic information. In this paper, we propose a discriminative embedding autoencoder with a regressor feedback model for ZSL. The encoder learns a mapping from the image feature space to the discriminative embedding space, which regulates both inter-class and intra-class distances between the learned features by a margin, making the learned features be discriminative for object recognition. The regressor feedback learns to map the reconstructed samples back to the the discriminative embedding and the semantic embedding, assisting the decoder to improve the quality of the samples and provide a generalization to the unseen classes. The proposed model is validated extensively on four benchmark datasets: SUN, CUB, AWA1, AWA2, the experiment results show that our proposed model outperforms the state-of-the-art models, and especially in the generalized zero-shot learning (GZSL), significant improvements are achieved.

cs.CV↗

arXiv2019

$\mathbb{Z}_\mathcal{N}$ graded discrete integrable systems and Darboux transformations

We present the Darboux transformations for a novel class of two-dimensional discrete integrable systems named as $\mathbb{Z}_\mathcal{N}$ graded discrete integrable systems, which were firstly proposed by Fordy and Xenitidis within the framework of $\mathbb{Z}_\mathcal{N}$ graded discrete Lax pairs very recently. In this paper, the $\mathbb{Z}_\mathcal{N}$ graded discrete equations in coprime case and their corresponding Lax pairs are derived from the discrete Gel'fand-Dikii hierarchy by applying a transformation of the independent variables. The construction of the Darboux tranformations is realised by considering the associated linear problems in the bilinear formalism for the $\mathbb{Z}_\mathcal{N}$ graded lattice equations. We show that all these $\mathbb{Z}_\mathcal{N}$ graded equations share a unified solution structure in our scheme.

nlin.SI↗

arXiv2019

Gaussian-Constrained training for speaker verification

Neural models, in particular the d-vector and x-vector architectures, have produced state-of-the-art performance on many speaker verification tasks. However, two potential problems of these neural models deserve more investigation. Firstly, both models suffer from `information leak', which means that some parameters participating in model training will be discarded during inference, i.e, the layers that are used as the classifier. Secondly, these models do not regulate the distribution of the derived speaker vectors. This `unconstrained distribution' may degrade the performance of the subsequent scoring component, e.g., PLDA. This paper proposes a Gaussian-constrained training approach that (1) discards the parametric classifier, and (2) enforces the distribution of the derived speaker vectors to be Gaussian. Our experiments on the VoxCeleb and SITW databases demonstrated that this new training approach produced more representative and regular speaker embeddings, leading to consistent performance improvement.

eess.AS↗

arXiv2018

Phonetic-attention scoring for deep speaker features in speaker verification

Recent studies have shown that frame-level deep speaker features can be derived from a deep neural network with the training target set to discriminate speakers by a short speech segment. By pooling the frame-level features, utterance-level representations, called d-vectors, can be derived and used in the automatic speaker verification (ASV) task. This simple average pooling, however, is inherently sensitive to the phonetic content of the utterance. An interesting idea borrowed from machine translation is the attention-based mechanism, where the contribution of an input word to the translation at a particular time is weighted by an attention score. This score reflects the relevance of the input word and the present translation. We can use the same idea to align utterances with different phonetic contents. This paper proposes a phonetic-attention scoring approach for d-vector systems. By this approach, an attention score is computed for each frame pair. This score reflects the similarity of the two frames in phonetic content, and is used to weigh the contribution of this frame pair in the utterance-based scoring. This new scoring approach emphasizes the frame pairs with similar phonetic contents, which essentially provides a soft alignment for utterances with any phonetic contents. Experimental results show that compared with the naive average pooling, this phonetic-attention scoring approach can deliver consistent performance improvement in ASV tasks of both text-dependent and text-independent.

eess.AS↗

arXiv2018

Exploring the Node Importance Based on von Neumann Entropy

When analyzing the statistical and topological characteristics of complex networks, an effective and convenient way is to compute the centralities for recognizing influential and significant nodes or structures, yet most of them are restricted to local environment or some specific configurations. In this paper we propose a new centrality for nodes based on the von Neumann entropy, which allows us to investigate the importance of nodes in the view of spectrum eigenvalues distribution. By presenting the performances of this centrality with network examples in reality, it is shown that the von Neumann entropy node centrality is an excellent index for selecting crucial nodes as well as classical ones. Then to lower down the computational complexity, an approximation calculation to this centrality is given which only depends on its first and second neighbors. Furthermore, in the optimal spreader problem and reducing average clustering coefficients, this entropy centrality presents excellent efficiency and unveil topological structure features of networks accurately. The entropy centrality could reduce the scales of giant connected components fastly in Erdos-Renyi and scale-free networks, and break down the cluster structures efficiently in random geometric graphs. This new methodology reveals the node importance in the perspective of spectrum, which provides a new insight into networks research and performs great potentials to discover essential structural features in networks.

cs.SI↗

arXiv2018

Binary Darboux transformations for discrete modified Boussinesq equation

We obtain the well-known discrete modified Boussinesq equation in two-component form as well as its Lax pair in $3\times3$ matrix form through a 3-periodic reduction technique on the Hirota-Miwa equation and its Lax pair. We describe how Darboux transformations and binary Darboux transformations can be constructed for this two-component discrete integrable equation. These transformations are then used to obtain classes of explicit solutions in the form of Casorati- and Gramm-type determinants. $N$-soliton solutions of the discrete modified Boussinesq equation are discussed as well when taking the vacuum potentials as constants.

nlin.SI↗

arXiv2018

Deep factorization for speech signal

Various informative factors mixed in speech signals, leading to great difficulty when decoding any of the factors. An intuitive idea is to factorize each speech frame into individual informative factors, though it turns out to be highly difficult. Recently, we found that speaker traits, which were assumed to be long-term distributional properties, are actually short-time patterns, and can be learned by a carefully designed deep neural network (DNN). This discovery motivated a cascade deep factorization (CDF) framework that will be presented in this paper. The proposed framework infers speech factors in a sequential way, where factors previously inferred are used as conditional variables when inferring other factors. We will show that this approach can effectively factorize speech signals, and using these factors, the original speech spectrum can be recovered with a high accuracy. This factorization and reconstruction approach provides potential values for many speech processing tasks, e.g., speaker recognition and emotion recognition, as will be demonstrated in the paper.

eess.AS↗

arXiv2017

Deep Factorization for Speech Signal

Speech signals are complex intermingling of various informative factors, and this information blending makes decoding any of the individual factors extremely difficult. A natural idea is to factorize each speech frame into independent factors, though it turns out to be even more difficult than decoding each individual factor. A major encumbrance is that the speaker trait, a major factor in speech signals, has been suspected to be a long-term distributional pattern and so not identifiable at the frame level. In this paper, we demonstrated that the speaker factor is also a short-time spectral pattern and can be largely identified with just a few frames using a simple deep neural network (DNN). This discovery motivated a cascade deep factorization (CDF) framework that infers speech factors in a sequential way, and factors previously inferred are used as conditional variables when inferring other factors. Our experiment on an automatic emotion recognition (AER) task demonstrated that this approach can effectively factorize speech signals, and using these factors, the original speech spectrum can be recovered with high accuracy. This factorization and reconstruction approach provides a novel tool for many speech processing tasks.

cs.SD↗

arXiv2017

Darboux and Binary Darboux Transformations for Discrete Integrable Systems. II. Discrete Potential mKdV Equation

The paper presents two results. First it is shown how the discrete potential modified KdV equation and its Lax pairs in matrix form arise from the Hirota-Miwa equation by a 2-periodic reduction. Then Darboux transformations and binary Darboux transformations are derived for the discrete potential modified KdV equation and it is shown how these may be used to construct exact solutions.

nlin.SI↗

arXiv2017

Phone-aware Neural Language Identification

Pure acoustic neural models, particularly the LSTM-RNN model, have shown great potential in language identification (LID). However, the phonetic information has been largely overlooked by most of existing neural LID models, although this information has been used in the conventional phonetic LID systems with a great success. We present a phone-aware neural LID architecture, which is a deep LSTM-RNN LID system but accepts output from an RNN-based ASR system. By utilizing the phonetic knowledge, the LID performance can be significantly improved. Interestingly, even if the test language is not involved in the ASR training, the phonetic knowledge still presents a large contribution. Our experiments conducted on four languages within the Babel corpus demonstrated that the phone-aware approach is highly effective.

cs.CL↗

arXiv2017

Deep Speaker Feature Learning for Text-independent Speaker Verification

Recently deep neural networks (DNNs) have been used to learn speaker features. However, the quality of the learned features is not sufficiently good, so a complex back-end model, either neural or probabilistic, has to be used to address the residual uncertainty when applied to speaker verification, just as with raw features. This paper presents a convolutional time-delay deep neural network structure (CT-DNN) for speaker feature learning. Our experimental results on the Fisher database demonstrated that this CT-DNN can produce high-quality speaker features: even with a single feature (0.3 seconds including the context), the EER can be as low as 7.68%. This effectively confirmed that the speaker trait is largely a deterministic short-time property rather than a long-time distributional pattern, and therefore can be extracted from just dozens of frames.

cs.SD↗

arXiv2017

Memory Visualization for Gated Recurrent Neural Networks in Speech Recognition

Recurrent neural networks (RNNs) have shown clear superiority in sequence modeling, particularly the ones with gated units, such as long short-term memory (LSTM) and gated recurrent unit (GRU). However, the dynamic properties behind the remarkable performance remain unclear in many applications, e.g., automatic speech recognition (ASR). This paper employs visualization techniques to study the behavior of LSTM and GRU when performing speech recognition tasks. Our experiments show some interesting patterns in the gated memory, and some of them have inspired simple yet effective modifications on the network structure. We report two of such modifications: (1) lazy cell update in LSTM, and (2) shortcut connections for residual learning. Both modifications lead to more comprehensible and powerful networks.

cs.LG↗