arXiv subjects

Bo Xu

Publications and source records attributed to Bo Xu.

At least 235 records · Page 13Linked to original sources

arXiv2021

Exploring wav2vec 2.0 on speaker verification and language identification

Wav2vec 2.0 is a recently proposed self-supervised framework for speech representation learning. It follows a two-stage training process of pre-training and fine-tuning, and performs well in speech recognition tasks especially ultra-low resource cases. In this work, we attempt to extend self-supervised framework to speaker verification and language identification. First, we use some preliminary experiments to indicate that wav2vec 2.0 can capture the information about the speaker and language. Then we demonstrate the effectiveness of wav2vec 2.0 on the two tasks respectively. For speaker verification, we obtain a new state-of-the-art result, Equal Error Rate (EER) of 3.61% on the VoxCeleb1 dataset. For language identification, we obtain an EER of 12.02% on 1 second condition and an EER of 3.47% on full-length condition of the AP17-OLR dataset. Finally, we utilize one model to achieve the unified modeling by the multi-task learning for the two tasks.

cs.SD↗

arXiv2021

Research on Fast Text Recognition Method for Financial Ticket Image

Currently, deep learning methods have been widely applied in and thus promoted the development of different fields. In the financial accounting field, the rapid increase in the number of financial tickets dramatically increases labor costs; hence, using a deep learning method to relieve the pressure on accounting is necessary. At present, a few works have applied deep learning methods to financial ticket recognition. However, first, their approaches only cover a few types of tickets. In addition, the precision and speed of their recognition models cannot meet the requirements of practical financial accounting systems. Moreover, none of the methods provides a detailed analysis of both the types and content of tickets. Therefore, this paper first analyzes the different features of 482 kinds of financial tickets, divides all kinds of financial tickets into three categories and proposes different recognition patterns for each category. These recognition patterns can meet almost all types of financial ticket recognition needs. Second, regarding the fixed format types of financial tickets (accounting for 68.27\% of the total types of tickets), we propose a simple yet efficient network named the Financial Ticket Faster Detection network (FTFDNet) based on a Faster RCNN. Furthermore, according to the characteristics of the financial ticket text, in order to obtain higher recognition accuracy, the loss function, Region Proposal Network (RPN), and Non-Maximum Suppression (NMS) are improved to make FTFDNet focus more on text. Finally, we perform a comparison with the best ticket recognition model from the ICDAR2019 invoice competition. The experimental results illustrate that FTFDNet increases the processing speed by 50\% while maintaining similar precision.

cs.CV↗

arXiv2020

Research on All-content Text Recognition Method for Financial Ticket Image

With the development of the economy, the number of financial tickets increases rapidly. The traditional manual invoice reimbursement and financial accounting system bring more and more burden to financial accountants. Therefore, based on the research and analysis of a large number of real financial ticket data, we designed an accurate and efficient all contents text detection and recognition method based on deep learning. This method has higher recognition accuracy and recall rate and can meet the actual requirements of financial accounting work. In addition, we propose a Financial Ticket Character Recognition Framework (FTCRF). According to the characteristics of Chinese character recognition, this framework contains a two-step information extraction method, which can improve the speed of Chinese character recognition. The experimental results show that the average recognition accuracy of this method is 91.75\% for character sequence and 87\% for the whole ticket. The availability and effectiveness of this method are verified by a commercial application system, which significantly improves the efficiency of the financial accounting system.

cs.CV↗

arXiv2020

Reference-Based Sequence Classification

Sequence classification is an important data mining task in many real world applications. Over the past few decades, many sequence classification methods have been proposed from different aspects. In particular, the pattern-based method is one of the most important and widely studied sequence classification methods in the literature. In this paper, we present a reference-based sequence classification framework, which can unify existing pattern-based sequence classification methods under the same umbrella. More importantly, this framework can be used as a general platform for developing new sequence classification algorithms. By utilizing this framework as a tool, we propose new sequence classification algorithms that are quite different from existing solutions. Experimental results show that new methods developed under the proposed framework are capable of achieving comparable classification accuracy to those state-of-the-art sequence classification algorithms.

cs.LG↗

arXiv2020

Audio-visual Speech Separation with Adversarially Disentangled Visual Representation

Speech separation aims to separate individual voice from an audio mixture of multiple simultaneous talkers. Although audio-only approaches achieve satisfactory performance, they build on a strategy to handle the predefined conditions, limiting their application in the complex auditory scene. Towards the cocktail party problem, we propose a novel audio-visual speech separation model. In our model, we use the face detector to detect the number of speakers in the scene and use visual information to avoid the permutation problem. To improve our model's generalization ability to unknown speakers, we extract speech-related visual features from visual inputs explicitly by the adversarially disentangled method, and use this feature to assist speech separation. Besides, the time-domain approach is adopted, which could avoid the phase reconstruction problem existing in the time-frequency domain models. To compare our model's performance with other models, we create two benchmark datasets of 2-speaker mixture from GRID and TCDTIMIT audio-visual datasets. Through a series of experiments, our proposed model is shown to outperform the state-of-the-art audio-only model and three audio-visual models.

cs.SD↗

arXiv2020

High voltage transition metal-free cathode material LiBC3F4 for Li ion batteries

The structural stability and electrochemical performance of boron substituted fluorinated graphite as a Li ion batteries cathode material are studied by first principles calculations. The results show that boron substituted fluorinated graphite BC3F4 possesses excellent structural stability, good electrical and ionic conductivities. Unexpectedly, the average Li intercalation voltage of LiBC3F4 is up to 4.44 V, which is much larger than that of LiBCF2. The average voltage of LiBC3F4 is even larger than that of common commercial transition metal oxides cathodes, indicating that LiBC3F4 is a breakthrough of transition metal-free high voltage cathode materials. By comparing the Fermi level, we found the Fermi level of LiBC3F4 is 1.38 eV lower than that of LiBCF2, leading to the decrease of the electron filling energy for the Li intercalation and forming much higher voltage. Moreover, LiBC3F4 shows small volume expansion and high energy density. LiBC3F4 is a promising high voltage cathode material for Li ion batteries. Finally, by calculating the evolution of Li intercalation voltage and Fermi level during the discharging process, a linear correlation between the Fermi level and Li intercalation voltage has been found.

cond-mat.mtrl-sci↗

arXiv2020

Financial ticket intelligent recognition system based on deep learning

Facing the rapid growth in the issuance of financial tickets (or bills, invoices etc.), traditional manual invoice reimbursement and financial accounting system are imposing an increasing burden on financial accountants and consuming excessive manpower. To solve this problem, we proposes an iterative self-learning Framework of Financial Ticket intelligent Recognition System (FFTRS), which can support the fast iterative updating and extensibility of the algorithm model, which are the fundamental requirements for a practical financial accounting system. In addition, we designed a simple yet efficient Financial Ticket Faster Detection network (FTFDNet) and an intelligent data warehouse of financial ticket are designed to strengthen its efficiency and performance. At present, the system can recognize 194 kinds of financial tickets and has an automatic iterative optimization mechanism, which means, with the increase of application time, the types of tickets supported by the system will continue to increase, and the accuracy of recognition will continue to improve. Experimental results show that the average recognition accuracy of the system is 97.07%, and the average running time for a single ticket is 175.67ms. The practical value of the system has been tested in a commercial application, which makes a beneficial attempt for the deep learning technology in financial accounting work.

cs.LG↗

arXiv2020

Finite Meta-Dynamic Neurons in Spiking Neural Networks for Spatio-temporal Learning

Spiking Neural Networks (SNNs) have incorporated more biologically-plausible structures and learning principles, hence are playing critical roles in bridging the gap between artificial and natural neural networks. The spikes are the sparse signals describing the above-threshold event-based firing and under-threshold dynamic computation of membrane potentials, which give us an alternative uniformed and efficient way on both information representation and computation. Inspired from the biological network, where a finite number of meta neurons integrated together for various of cognitive functions, we proposed and constructed Meta-Dynamic Neurons (MDN) to improve SNNs for a better network generalization during spatio-temporal learning. The MDNs are designed with basic neuronal dynamics containing 1st-order and 2nd-order dynamics of membrane potentials, including the spatial and temporal meta types supported by some hyper-parameters. The MDNs generated from a spatial (MNIST) and a temporal (TIDigits) datasets first, and then extended to various other different spatio-temporal tasks (including Fashion-MNIST, NETtalk, Cifar-10, TIMIT and N-MNIST). The comparable accuracy was reached compared to other SOTA SNN algorithms, and a better generalization was also achieved by SNNs using MDNs than that without using MDNs.

cs.NE↗

arXiv2020

Temperature-dependent hardness of diamond-structured covalent materials

Understanding temperature-dependent hardness of covalent materials is not only of fundamental scientific interest, but also of crucial importance for technical applications. In this work, a temperature-dependent hardness formula for diamond-structured covalent materials is constructed on the basis of the dislocation theory. Our results show that, at low temperature, the Vickers hardness is mainly controlled by Poisson's ratio and shear modulus with the latter playing a dominant role. With increasing temperature, the plastic deformation mechanism undergoes a transition from shuffle-set dislocation control to glide-set dislocation control, leading to a steeper drop of hardness at high temperature. In addition, an intrinsic parameter, a3G, is revealed for diamond-structured covalent materials, which measures the resistance to soften at high temperature. Our hardness model shows remarkable agreement with experimental data. Current work not only sheds lights on the physical origin of hardness, but also provides a direct principle for superhard materials design.

cond-mat.mtrl-sci↗

arXiv2020

Shifu2: A Network Representation Learning Based Model for Advisor-advisee Relationship Mining

The advisor-advisee relationship represents direct knowledge heritage, and such relationship may not be readily available from academic libraries and search engines. This work aims to discover advisor-advisee relationships hidden behind scientific collaboration networks. For this purpose, we propose a novel model based on Network Representation Learning (NRL), namely Shifu2, which takes the collaboration network as input and the identified advisor-advisee relationship as output. In contrast to existing NRL models, Shifu2 considers not only the network structure but also the semantic information of nodes and edges. Shifu2 encodes nodes and edges into low-dimensional vectors respectively, both of which are then utilized to identify advisor-advisee relationships. Experimental results illustrate improved stability and effectiveness of the proposed model over state-of-the-art methods. In addition, we generate a large-scale academic genealogy dataset by taking advantage of Shifu2.

cs.LG↗

arXiv2020

Spin-injection and spin-relaxation in p-doped InGaAs/GaAs quantum-dot spin light emitting diode at zero magnetic field

We report on efficient spin injection in p-doped InGaAs/GaAs quantum-dot (QD) spin light emitting diode (spin-LED) under zero applied magnetic field. A high degree of electroluminescence circular polarization (Pc) ~19% is measured in remanence up to 100K. This result is obtained thanks to the combination of a perpendicularly magnetized CoFeB/MgO spin injector allowing efficient spin injection and an appropriate p-doped InGaAs/GaAs QD layer in the active region. By analyzing the bias and temperature dependence of the electroluminescence circular polarization, we have evidenced a two-step spin relaxation process. The first step occurs when electrons tunnel through the MgO barrier and travel across the GaAs depletion layer. The spin relaxation is dominated by the Dyakonov-Perel mechanism related to the kinetic energy of electrons, which is characterized by a bias dependent Pc. The second step occurs when electrons are captured into QDs prior to their radiative recombination with holes. The temperature dependence of Pc reflects the temperature induced modification of the QDs doping, together with the variation of the ratio between the charge carrier lifetime and the spin relaxation time inside the QDs. The understanding of these spin relaxation mechanisms is essential to improve the performance of spin LED for future spin optoelectronic applications at room temperature under zero applied magnetic field.

cond-mat.mtrl-sci↗

arXiv2020

DINE: A Framework for Deep Incomplete Network Embedding

Network representation learning (NRL) plays a vital role in a variety of tasks such as node classification and link prediction. It aims to learn low-dimensional vector representations for nodes based on network structures or node attributes. While embedding techniques on complete networks have been intensively studied, in real-world applications, it is still a challenging task to collect complete networks. To bridge the gap, in this paper, we propose a Deep Incomplete Network Embedding method, namely DINE. Specifically, we first complete the missing part including both nodes and edges in a partially observable network by using the expectation-maximization framework. To improve the embedding performance, we consider both network structures and node attributes to learn node representations. Empirically, we evaluate DINE over three networks on multi-label classification and link prediction tasks. The results demonstrate the superiority of our proposed approach compared against state-of-the-art baselines.

cs.SI↗

arXiv2020

MODEL: Motif-based Deep Feature Learning for Link Prediction

Link prediction plays an important role in network analysis and applications. Recently, approaches for link prediction have evolved from traditional similarity-based algorithms into embedding-based algorithms. However, most existing approaches fail to exploit the fact that real-world networks are different from random networks. In particular, real-world networks are known to contain motifs, natural network building blocks reflecting the underlying network-generating processes. In this paper, we propose a novel embedding algorithm that incorporates network motifs to capture higher-order structures in the network. To evaluate its effectiveness for link prediction, experiments were conducted on three types of networks: social networks, biological networks, and academic networks. The results demonstrate that our algorithm outperforms both the traditional similarity-based algorithms by 20% and the state-of-the-art embedding-based algorithms by 19%.

cs.SI↗

arXiv2020

A spectrally bright wavelength-switchable vacuum ultraviolet source driven by quantum coherence in strong-field-ionized molecules

We report generation of spectrally bright vacuum ultraviolet (VUV) and deep UV (DUV) coherent radiations driven by quantum coherence in tunnel-ionized carbon monoxide (CO) molecules. Our technique allows us to switch between multiple wavelengths provided by the abundant energy levels of molecular ions. The DUV/VUV sources can have arbitrary polarization states by manipulating the pump laser polarization. The superior temporal and spectral properties of the developed source give rise to a broadband Raman comb in the DUV/VUV region.

physics.optics↗

arXiv2020

Electrical Detection of Light Helicity using a Quantum Dots based Hybrid Device at Zero Magnetic Field

Photon helicity-dependent photocurrent is measured at zero magnetic field on a device based on an ensemble of InGaAs/GaAs quantum dots that are embedded into a GaAs-based p-i-n diode. Our main goal is to take advantage of the long electron spin relaxation time expected in these nano-objects. In these experiments, no external magnetic field is required thanks to the use of an ultrathin magnetic CoFeB/MgO electrode, presenting perpendicular magnetic anisotropy (PMA). We observe a clear asymmetry of the photocurrent measured under respective right and left polarized light that follows the hysteresis of the magnetic layer. The amplitude of this asymmetry at zero magnetic field decreases with increasing temperatures and can be controlled with the bias. Polarization-resolved photoluminescence is detected in parallel while the device is operated as a photodetector. This demonstrates the multifunctional capabilities of the device and gives valuable insights into the spin relaxation of the electrons in the quantum dots.

physics.app-ph↗

arXiv2020

Speaker-Conditional Chain Model for Speech Separation and Extraction

Speech separation has been extensively explored to tackle the cocktail party problem. However, these studies are still far from having enough generalization capabilities for real scenarios. In this work, we raise a common strategy named Speaker-Conditional Chain Model to process complex speech recordings. In the proposed method, our model first infers the identities of variable numbers of speakers from the observation based on a sequence-to-sequence model. Then, it takes the information from the inferred speakers as conditions to extract their speech sources. With the predicted speaker information from whole observation, our model is helpful to solve the problem of conventional speech separation and speaker extraction for multi-round long recordings. The experiments from standard fully-overlapped speech separation benchmarks show comparable results with prior studies, while our proposed model gets better adaptability for multi-round long recordings.

eess.AS↗

arXiv2020

Sequence to Multi-Sequence Learning via Conditional Chain Mapping for Mixture Signals

Neural sequence-to-sequence models are well established for applications which can be cast as mapping a single input sequence into a single output sequence. In this work, we focus on one-to-many sequence transduction problems, such as extracting multiple sequential sources from a mixture sequence. We extend the standard sequence-to-sequence model to a conditional multi-sequence model, which explicitly models the relevance between multiple output sequences with the probabilistic chain rule. Based on this extension, our model can conditionally infer output sequences one-by-one by making use of both input and previously-estimated contextual output sequences. This model additionally has a simple and efficient stop criterion for the end of the transduction, making it able to infer the variable number of output sequences. We take speech data as a primary test field to evaluate our methods since the observed speech data is often composed of multiple sources due to the nature of the superposition principle of sound waves. Experiments on several different tasks including speech separation and multi-speaker speech recognition show that our conditional multi-sequence models lead to consistent improvements over the conventional non-conditional models.

eess.AS↗

arXiv2020

A Comparison of Label-Synchronous and Frame-Synchronous End-to-End Models for Speech Recognition

End-to-end models are gaining wider attention in the field of automatic speech recognition (ASR). One of their advantages is the simplicity of building that directly recognizes the speech frame sequence into the text label sequence by neural networks. According to the driving end in the recognition process, end-to-end ASR models could be categorized into two types: label-synchronous and frame-synchronous, each of which has unique model behaviour and characteristic. In this work, we make a detailed comparison on a representative label-synchronous model (transformer) and a soft frame-synchronous model (continuous integrate-and-fire (CIF) based model). The results on three public dataset and a large-scale dataset with 12000 hours of training data show that the two types of models have respective advantages that are consistent with their synchronous mode.

eess.AS↗