arXiv subjects

Dong Wang

Publications and source records attributed to Dong Wang.

At least 721 records · Page 40Linked to original sources

arXiv2017

Phonetic Temporal Neural Model for Language Identification

Deep neural models, particularly the LSTM-RNN model, have shown great potential for language identification (LID). However, the use of phonetic information has been largely overlooked by most existing neural LID methods, although this information has been used very successfully in conventional phonetic LID systems. We present a phonetic temporal neural model for LID, which is an LSTM-RNN LID system that accepts phonetic features produced by a phone-discriminative DNN as the input, rather than raw acoustic features. This new model is similar to traditional phonetic LID methods, but the phonetic knowledge here is much richer: it is at the frame level and involves compacted information of all phones. Our experiments conducted on the Babel database and the AP16-OLR database demonstrate that the temporal phonetic neural approach is very effective, and significantly outperforms existing acoustic neural models. It also outperforms the conventional i-vector approach on short utterances and in noisy conditions.

cs.CL↗

arXiv2017

Design of a Nondestructive Two-in-One Instrument for Measuring the Polarization and Energy Spectrum at an X-ray FEL Facility

The free electron laser (FEL), as a next-generation light source, is an attractive tool in scientific frontier research because of its advantages of full coherence, ultra-short pulse duration, and controllable polarization. Owing to the demand of real-time bunch diagnosis during FEL experiments, precise nondestructive measurements of the polarization and X-ray energy spectrum using one instrument are preferred. In this paper, such an instrument based on the electron time-of-flight technique is proposed. By considering the complexity and nonlinearity, a numerical model in the framework of Geant4 has been developed for optimization. Taking the Shanghai Soft X-ray FEL user facility as an example, its measurement performances' dependencies on the critical parameters were studied systematically, and, finally, an optimal design was obtained, achieving resolutions of 0.5% for the polarization degree and 0.3 eV for the X-ray energy spectrum.

physics.ins-det↗

arXiv2017

Amulet: Aggregating Multi-level Convolutional Features for Salient Object Detection

Fully convolutional neural networks (FCNs) have shown outstanding performance in many dense labeling problems. One key pillar of these successes is mining relevant information from features in convolutional layers. However, how to better aggregate multi-level convolutional feature maps for salient object detection is underexplored. In this work, we present Amulet, a generic aggregating multi-level convolutional feature framework for salient object detection. Our framework first integrates multi-level feature maps into multiple resolutions, which simultaneously incorporate coarse semantics and fine details. Then it adaptively learns to combine these feature maps at each resolution and predict saliency maps with the combined features. Finally, the predicted results are efficiently fused to generate the final saliency map. In addition, to achieve accurate boundary inference and semantic enhancement, edge-aware feature maps in low-level layers and the predicted results of low resolution features are recursively embedded into the learning framework. By aggregating multi-level convolutional features in this efficient and flexible manner, the proposed saliency model provides accurate salient object labeling. Comprehensive experiments demonstrate that our method performs favorably against state-of-the art approaches in terms of near all compared evaluation metrics.

cs.CV↗

arXiv2017

Memory-augmented Neural Machine Translation

Neural machine translation (NMT) has achieved notable success in recent times, however it is also widely recognized that this approach has limitations with handling infrequent words and word pairs. This paper presents a novel memory-augmented NMT (M-NMT) architecture, which stores knowledge about how words (usually infrequently encountered ones) should be translated in a memory and then utilizes them to assist the neural model. We use this memory mechanism to combine the knowledge learned from a conventional statistical machine translation system and the rules learned by an NMT system, and also propose a solution for out-of-vocabulary (OOV) words based on this framework. Our experiments on two Chinese-English translation tasks demonstrated that the M-NMT architecture outperformed the NMT baseline by $9.0$ and $2.7$ BLEU points on the two tasks, respectively. Additionally, we found this architecture resulted in a much more effective OOV treatment compared to competitive methods.

cs.CL↗

arXiv2017

Learning Uncertain Convolutional Features for Accurate Saliency Detection

Deep convolutional neural networks (CNNs) have delivered superior performance in many computer vision tasks. In this paper, we propose a novel deep fully convolutional network model for accurate salient object detection. The key contribution of this work is to learn deep uncertain convolutional features (UCF), which encourage the robustness and accuracy of saliency detection. We achieve this via introducing a reformulated dropout (R-dropout) after specific convolutional layers to construct an uncertain ensemble of internal feature units. In addition, we propose an effective hybrid upsampling method to reduce the checkerboard artifacts of deconvolution operators in our decoder network. The proposed methods can also be applied to other deep convolutional networks. Compared with existing saliency detection methods, the proposed UCF model is able to incorporate uncertainties for more accurate object boundary inference. Extensive experiments demonstrate that our proposed saliency model performs favorably against state-of-the-art approaches. The uncertain feature learning mechanism as well as the upsampling method can significantly improve performance on other pixel-wise vision tasks.

cs.CV↗

arXiv2017

How the relativistic motion affects Einstein-Podolsky-Rosen steering

In this Letter, the dynamic behavior of Einstein-Podolsky-Rosen (EPR) steering and the redistribution of EPR steering under a relativistic framework are investigated. Specifically, we explore the scenario that particle A hold by Alice is in a flat space-time and another particle B by Bob entangled with A is in a non-inertial framework. The results show that EPR steering from Alice to Bob is dramatically destroyed by Unruh effect caused by the acceleration of Bob. Besides, EPR steering has an asymmetry property, and EPR steering asymmetry increases with the growing intensity of Unruh effect, implying that the Unruh effect can bring on EPR steering asymmetry. Furthermore, the reduced physical accessible EPR steering from Alice to Bob is distributed to the physical inaccessible EPR steering (from Alice to anti-Bob or from Bob to anti-Bob). Notably, unlike entanglement and quantum discord, only one of EPR steering from Alice to anti-Bob and Bob to anti-Bob experiences a sudden birth with the increase of acceleration parameter, which means that they cannot simultaneously survive. That is, the monogamy relation of EPR steering is still tenable in such a scenario. Consequently, we believe that EPR steering could also be served as one of important information resources within the long-distance quantum secure communication under the relativistic framework.

quant-ph↗

arXiv2017

AP17-OLR Challenge: Data, Plan, and Baseline

We present the data profile and the evaluation plan of the second oriental language recognition (OLR) challenge AP17-OLR. Compared to the event last year (AP16-OLR), the new challenge involves more languages and focuses more on short utterances. The data is offered by SpeechOcean and the NSFC M2ASR project. Two types of baselines are constructed to assist the participants, one is based on the i-vector model and the other is based on various neural networks. We report the baseline results evaluated with various metrics defined by the AP17-OLR evaluation plan and demonstrate that the combined database is a reasonable data resource for multilingual research. All the data is free for participants, and the Kaldi recipes for the baselines have been published online.

cs.CL↗

arXiv2017

Memory-augmented Chinese-Uyghur Neural Machine Translation

Neural machine translation (NMT) has achieved notable performance recently. However, this approach has not been widely applied to the translation task between Chinese and Uyghur, partly due to the limited parallel data resource and the large proportion of rare words caused by the agglutinative nature of Uyghur. In this paper, we collect ~200,000 sentence pairs and show that with this middle-scale database, an attention-based NMT can perform very well on Chinese-Uyghur/Uyghur-Chinese translation. To tackle rare words, we propose a novel memory structure to assist the NMT inference. Our experiments demonstrated that the memory-augmented NMT (M-NMT) outperforms both the vanilla NMT and the phrase-based statistical machine translation (SMT). Interestingly, the memory structure provides an elegant way for dealing with words that are out of vocabulary.

cs.CL↗

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

Deep Speaker Verification: Do We Need End to End?

End-to-end learning treats the entire system as a whole adaptable black box, which, if sufficient data are available, may learn a system that works very well for the target task. This principle has recently been applied to several prototype research on speaker verification (SV), where the feature learning and classifier are learned together with an objective function that is consistent with the evaluation metric. An opposite approach to end-to-end is feature learning, which firstly trains a feature learning model, and then constructs a back-end classifier separately to perform SV. Recently, both approaches achieved significant performance gains on SV, mainly attributed to the smart utilization of deep neural networks. However, the two approaches have not been carefully compared, and their respective advantages have not been well discussed. In this paper, we compare the end-to-end and feature learning approaches on a text-independent SV task. Our experiments on a dataset sampled from the Fisher database and involving 5,000 speakers demonstrated that the feature learning approach outperformed the end-to-end approach. This is a strong support for the feature learning approach, at least with data and computation resources similar to ours.

cs.SD↗

arXiv2017

Speaker Recognition with Cough, Laugh and "Wei"

This paper proposes a speaker recognition (SRE) task with trivial speech events, such as cough and laugh. These trivial events are ubiquitous in conversations and less subjected to intentional change, therefore offering valuable particularities to discover the genuine speaker from disguised speech. However, trivial events are often short and idiocratic in spectral patterns, making SRE extremely difficult. Fortunately, we found a very powerful deep feature learning structure that can extract highly speaker-sensitive features. By employing this tool, we studied the SRE performance on three types of trivial events: cough, laugh and "Wei" (a short Chinese "Hello"). The results show that there is rich speaker information within these trivial events, even for cough that is intuitively less speaker distinguishable. With the deep feature approach, the EER can reach 10%-14% with the three trivial events, despite their extremely short durations (0.2-1.0 seconds).

cs.SD↗

arXiv2017

Cross-lingual Speaker Verification with Deep Feature Learning

Existing speaker verification (SV) systems often suffer from performance degradation if there is any language mismatch between model training, speaker enrollment, and test. A major cause of this degradation is that most existing SV methods rely on a probabilistic model to infer the speaker factor, so any significant change on the distribution of the speech signal will impact the inference. Recently, we proposed a deep learning model that can learn how to extract the speaker factor by a deep neural network (DNN). By this feature learning, an SV system can be constructed with a very simple back-end model. In this paper, we investigate the robustness of the feature-based SV system in situations with language mismatch. Our experiments were conducted on a complex cross-lingual scenario, where the model training was in English, and the enrollment and test were in Chinese or Uyghur. The experiments demonstrated that the feature-based system outperformed the i-vector system with a large margin, particularly with language mismatch between enrollment and test.

cs.SD↗

arXiv2017

Factor Models for Matrix-Valued High-Dimensional Time Series

In finance, economics and many other fields, observations in a matrix form are often observed over time. For example, many economic indicators are obtained in different countries over time. Various financial characteristics of many companies are reported over time. Although it is natural to turn a matrix observation into a long vector then use standard vector time series models or factor analysis, it is often the case that the columns and rows of a matrix represent different sets of information that are closely interrelated in a very structural way. We propose a novel factor model that maintains and utilizes the matrix structure to achieve greater dimensional reduction as well as finding clearer and more interpretable factor structures. Estimation procedure and its theoretical properties are investigated and demonstrated with simulated and real examples.

stat.ME↗

arXiv2017

Critical Height of the Torus Instability in Two-Ribbon Solar Flares

We studied the background field for 60 two-ribbon flares of M-and-above classes during 2011--2015. These flares are categorized into two groups, i.e., \emph{eruptive} and \emph{confined} flares, based on whether a flare is associated with a coronal mass ejection or not. The background field of source active regions is approximated by a potential field extrapolated from the $B_z$ component of vector magnetograms provided by the Helioseismic and Magnetic Imager. We calculated the decay index $n$ of the background field above the flaring polarity inversion line, and defined a critical height $h_\mathrm{crit}$ corresponding to the theoretical threshold ($n_\mathrm{crit}=1.5$) of the torus instability. We found that $h_\mathrm{crit}$ is approximately half of the distance between the centroids of opposite polarities in active regions, and that the distribution of $h_\mathrm{crit}$ is bimodal: it is significantly higher for confined flares than for eruptive ones. The decay index increases monotonously with increasing height for 86\% (84\%) of the eruptive (confined) flares but displays a saddle-like profile for the rest 14\% (16\%), which are found exclusively in active regions of multipolar field configuration. Moreover, $n$ at the saddle bottom is significantly smaller in confined flares than that in eruptive ones. These results highlight the critical role of background field in regulating the eruptive behavior of two-ribbon flares.

astro-ph.SR↗

arXiv2017

A Study on Replay Attack and Anti-Spoofing for Automatic Speaker Verification

For practical automatic speaker verification (ASV) systems, replay attack poses a true risk. By replaying a pre-recorded speech signal of the genuine speaker, ASV systems tend to be easily fooled. An effective replay detection method is therefore highly desirable. In this study, we investigate a major difficulty in replay detection: the over-fitting problem caused by variability factors in speech signal. An F-ratio probing tool is proposed and three variability factors are investigated using this tool: speaker identity, speech content and playback & recording device. The analysis shows that device is the most influential factor that contributes the highest over-fitting risk. A frequency warping approach is studied to alleviate the over-fitting problem, as verified on the ASV-spoof 2017 database.

cs.SD↗

arXiv2017

Weakly Supervised PLDA Training

PLDA is a popular normalization approach for the i-vector model, and it has delivered state-of-the-art performance in speaker verification. However, PLDA training requires a large amount of labelled development data, which is highly expensive in most cases. We present a cheap PLDA training approach, which assumes that speakers in the same session can be easily separated, and speakers in different sessions are simply different. This results in `weak labels' which are not fully accurate but cheap, leading to a weak PLDA training. Our experimental results on real-life large-scale telephony customer service achieves demonstrated that the weak training can offer good performance when human-labelled data are limited. More interestingly, the weak training can be employed as a discriminative adaptation approach, which is more efficient than the prevailing unsupervised method when human-labelled data are insufficient.

cs.LG↗

arXiv2017

Collaborative Learning for Language and Speaker Recognition

This paper presents a unified model to perform language and speaker recognition simultaneously and altogether. The model is based on a multi-task recurrent neural network where the output of one task is fed as the input of the other, leading to a collaborative learning framework that can improve both language and speaker recognition by borrowing information from each other. Our experiments demonstrated that the multi-task model outperforms the task-specific models on both tasks.

cs.SD↗

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↗