Searcharxiv⌕ Search

arXiv subjects

Ya Li

Publications and source records attributed to Ya Li.

At least 91 records · Page 5Linked to original sources
arXiv2022

ECAPA-TDNN for Multi-speaker Text-to-speech Synthesis

In recent years, neural network based methods for multi-speaker text-to-speech synthesis (TTS) have made significant progress. However, the current speaker encoder models used in these methods still cannot capture enough speaker information. In this paper, we focus on accurate speaker encoder modeling and propose an end-to-end method that can generate high-quality speech and better similarity for both seen and unseen speakers. The proposed architecture consists of three separately trained components: a speaker encoder based on the state-of-the-art ECAPA-TDNN model which is derived from speaker verification task, a FastSpeech2 based synthesizer, and a HiFi-GAN vocoder. The comparison among different speaker encoder models shows our proposed method can achieve better naturalness and similarity. To efficiently evaluate our synthesized speech, we are the first to adopt deep learning based automatic MOS evaluation methods to assess our results, and these methods show great potential in automatic speech quality assessment.

cs.SD↗
arXiv2022

$CP$-violating observables in four-body $B\rightarrow ϕ(\rightarrow K\bar K)K^*(\rightarrow Kπ)$ decays

We analyse the four-body $B\rightarrow ϕ(\rightarrow K\bar K)K^*(\rightarrow Kπ)$ decays in the perturbative QCD approach,where the invariant mass of $K\bar K$($Kπ$) system is limited in a window of $\pm 15$ MeV ($\pm150$ MeV) around the $ϕ(K^*(892))$ mass.In addition to the P wave resonances,two important S wave backgrounds in the selected invariant mass region are also accounted for. Angular momentum conservation allows six helicity amplitudes to contribute,including three P waves, two single S waves,and one double S wave. We calculated the branching ratio for each component and found sizable S wave contributions,coincide with the experimental observation.The obtained branching ratios of $B^{0(+)}\rightarrow ϕK^{*0(+)}$ are comparable with the previous predictions and support the measurements, whereas the predicted $\mathcal{B}(B^0_s\rightarrow ϕ\bar K^{*0})$ is smaller than the world average. The longitudinal polarizations are predicted to be around 0.7,consistent with previous PQCD results but larger than the data. Aside from the direct CP asymmetries,the true and fake triple product asymmetries(TPAs) are calculated in this work. In the case of neutral modes, both direct CP asymmetries and true TPAs are expected to be zero due to the vanishing weak phase difference. The direct CP asymmetries for the $B^+$ mode are predicted to be tiny,since the tree contributions are suppressed with respect to the penguin ones. The true asymmetries have shown no significant deviations from zero.In contrast,large fake asymmetries are observed in these decays,indicating the presence of significant final state interactions.We give the predictions of the S wave induced TPAs for the first time,which is consistent with LHCb data and would be checked with future measurements from Belle and BABAR experiments if the S wave components can be properly taken into account in angular analysis.

hep-ph↗
arXiv2021

Global determination of two-meson distribution amplitudes from three-body $B$ decays in the perturbative QCD approach

We perform a global analysis of three-body charmless hadronic decays $B\to VP_3\to P_1P_2P_3$ in the perturbative QCD (PQCD) approach, where $V$ denotes an intermediate vector resonance, and $P_i$, $i=1,2,3$, denote final-state pseudoscalar mesons. Fitting the PQCD factorization formulas at leading order in the strong coupling $α_s$ to measured branching ratios and direct $CP$ asymmetries, we determine the Gegenbauer moments in the two-meson distribution amplitudes (DAs) for the meson pairs $P_1P_2=ππ, Kπ, KK$. The fitted Gegenbauer moments are then employed to make predictions for those observables, whose data are excluded in the fit due to larger experimental uncertainties. A general consistency between our predictions and data is achieved, which hints the validity of the PQCD formalism for the above three-body $B$ meson decays and the universality of the nonperturbative two-meson DAs. The obtained two-meson DAs can be applied to PQCD studies of other multi-body $B$ meson decays involving the same meson pairs. We also attempt to determine the dependence of the Gegenbauer moments on the meson-pair invariant mass, and demonstrate that this determination is promising, when data become more precise.

hep-ph↗
arXiv2021

Noise suppression in a temporal-multimode quantum memory entangled with a photon via asymmetrical photon-collection channel

Quantum interfaces (QIs) that generate entanglement between a multimode atomic memory and a photon forms a multiplexed repeater node and hold promise to greatly improve quantum repeater rates. Recently, the temporal multimode spin-wave memory that is entangled with a photon has been demonstrated with cold atoms. However, due to additional noise generated in multimode operation, the fidelity of spin-wave-photon entanglement significantly decreases with the mode number. So far, the improvement on temporal-multimode entanglement fidelity via suppressing the additional noise remains unexplored. Here, we propose and experimentally demonstrate a scheme that can suppress the additional noise of a temporally-multiplexed QI. The scheme uses an asymmetric channel to collect the photons coming and retrieving from the temporally-multiplexed QI. For making comparisons, we also set up a QI that uses symmetric channel for the photon collections. When the QIs store 14 modes, the measured Bell parameter S for the QIs using the asymmetric and the symmetric photon-collection channels are 2.36+/-0.03 and 2.24+/-0.04, respectively, showing that the QI using the asymmetric channel gives rise to a 3% increase in entanglement fidelity, i.e., a 1.7-fold decrease in the additional noise, compared with the QI using the symmetric one. On the other hand, the 14-mode entanglement QIs that use the asymmetric and symmetric collections preserve the violation of a Bell inequality for storage times up to 25 us and 20 us, respectively, showing that the asymmetric QI has a higher entanglement storage performance.

quant-ph↗
arXiv2021

Generation of entanglement between a highly wave-packet-tunable photon and a spin-wave memory in cold atoms

Controls of waveforms (pulse durations) of single photons are important tasks for effectively interconnecting disparate atomic memories in hybrid quantum networks. So far, the waveform control of single photon that is entangled with an atomic memory remains unexplored. Here, we demonstrated control of waveform length of the photon that is entangled with an atomic spin-wave memory by varying light-atom interaction time in cold atoms. The Bell parameter S as a function of the duration of photon pulse is measured, which shows that violations of Bell equality can be achieved for the photon pulse in the duration range from 40 ns to 50 us, where, S=2.64+/-0.02 and S=2.26+/-0.05 for the 40-ns and 50-μs durations, respectively. The measured results show that S parameter decreases with the increase in the pulse duration. We confirm that the increase in photon noise probability per pulse with the pulse-duration is responsible for the S decrease.

quant-ph↗
arXiv2021

Generation of perfect-cavity-enhanced atom-photon entanglement with a millisecond lifetime via a spatially-multiplexed cavity

A qubit memory is the building block for quantum information. Cavity-enhanced spin-wave-photon entanglement has been achieved by applying dual-control modes. However, owing to cross readouts between the modes, the qubit retrieval efficiency is about one quarter lower than that for a single spin-wave mode at all storage times. Here, we overcome cross readouts using a multiplexed ring cavity. The cavity is embedded with a polarization interferometer, and we create a write-out photonic qubit entangled with a magnetic-field-insensitive spin-wave qubit by applying a single-mode write-laser beam to cold atoms. The spin-wave qubit is retrieved with a single-mode read-laser beam, and the quarter retrieval-efficiency loss is avoided at all storage times. Our experiment demonstrates 50% intrinsic retrieval efficiency for 540 microsecond storage time, which is 13.5 times longer than the best reported result. Importantly, our multiplexed-cavity scheme paves one road to generate perfect-cavity-enhanced and large-scale multiplexed spin-wave-photon entanglement with a long lifetime.

quant-ph↗
arXiv2021

Study of $B_{(s)} \to (ππ)(Kπ)$ decays in the perturbative QCD approach

In this work, we provide estimates of the branching ratios, direct $CP$ asymmetries and triple product asymmetries in $B_{(s)} \to (ππ)(Kπ)$ decays in the perturbative QCD approach, where the $ππ$ and $Kπ$ invariant mass spectra are dominated by the vector resonances $ρ(770)$ and $K^*(892)$, respectively. Some scalar backgrounds, such as $f_0(500,980) \to ππ$ and $K^*_0(1430) \to Kπ$ are also accounted for. The $ρ(700)$ is parametrized by the Gounaris-Sakurai function. The relativistic Breit-Wigner formula for the $f_0(500)$ and Flatté model for the $f_0(980)$ are adopted to parameterize the time-like scalar form factors $F_S(ω^2)$. We also use the D.V. Bugg model to parameterize the $f_0(500)$ and compare the relevant theoretical predictions from different models. While in the region of $Kπ$ invariant mass, the $K^*_0(1430)$ is described with the LASS lineshape and the $K^*(892)$ is modeled by the Breit-Wigner function. We find that the decay rates for the considered decay modes agree with currently available data within errors. As a by-product, we extract the branching ratios of two-body decays $B_{(s)} \to ρ(770)K^*(892)$ from the corresponding four-body decay modes and calculate the relevant polarization fractions. Our prediction of longitudinal polarization fraction for $B^0\to ρ(770)^0 K^*(892)^0$ decay deviates a lot from the recent LHCb measurement, which should be resolved. It is shown that the direct $CP$ asymmetries are large due to the sizable interference between the tree and penguin contributions, but they are small for the tree-dominant or penguin-dominant processes. The PQCD predictions for the triple product asymmetries are small which are expected in the standard model, and consistent with the current data reported by the LHCb Collaboration.Our results can be tested by the future precise data from the LHCb and Belle II experiments.

hep-ph↗
arXiv2021

Four-body decays $B_{(s)} \rightarrow (Kπ)_{S/P} (Kπ)_{S/P}$ in the perturbative QCD approach

We investigate the four-body decays $B_{(s)} \rightarrow (Kπ)_{S/P} (Kπ)_{S/P}$ in the $Kπ$-pair invariant mass region around the $K^*(892)$ resonance in the perturbative QCD (PQCD) approach, where $(Kπ)_{S/P}$ denotes a $S$- or $P$-wave $Kπ$ configuration. The contributions from the $P$-wave resonance $K^*(892)$ and from possible $S$-wave components are included into the parametrization of the $Kπ$ two-meson distribution amplitudes, which are the nonperturbative inputs in this formalism. We calculate the branching ratio for each resonance and observe sizable $S$-wave contributions to the $B_s$ modes, well consistent with recently available LHCb data. We also deduce the polarization fractions for the $K^*\bar{K}^*$ final states, and compare our predictions to those in previous PQCD analyses of the corresponding two-body $B\to K^*\bar{K}^*$ decays. The polarization puzzle associated with the two $U$-spin related channels $B^0\rightarrow K^{*0}\bar{K}^{*0}$ and $B^0_s\rightarrow K^{*0}\bar{K}^{*0}$ still exists. In addition to direct $CP$ asymmetries, triple-product asymmetries and $S$-wave induced direct $CP$ asymmetries originating from the interference among various helicity amplitudes, are presented. It is shown that true triple-product asymmetries are rather small, while direct $CP$ asymmetries and $S$-wave-induced direct $CP$ asymmetries are significant in some decays. Our results will be subject to stringent tests with precise data from $B$ factories in the near future.

hep-ph↗
arXiv2020

Resonant contributions to three-body $B_{(s)} \to [ D^{(*)}, \bar{D}^{(*)} ] K^+K^-$ decays in the perturbative QCD approach

In this work, we study the $S$, $P$ and $D$ wave resonance contributions to three-body decays $B_{(s)} \to [ D^{(*)}, \bar{D}^{(*)} ] K^+K^-$ by employing the perturbative QCD (PQCD) approach, where the kaon-kaon invariant mass spectra are dominated by the $f_0(980),f_0(1370),$ $ϕ(1020),ϕ(1680),f_2(1270),f^{\prime}_2(1525),f_2(1750)$ and $f_2(1950)$ resonances. The $KK$ $S$-wave component $f_0(980)$ is modeled with the Flatté formalism, while other resonances are described by the relativistic Breit-Wigner (BW) line shape. The corresponding decay channels are studied by constructing the kaon-kaon distribution amplitude $Φ_{KK}$, which captures important final state interactions between the kaon pair in the resonant region. We found that the PQCD predictions for the branching ratios for most considered decays agree with currently available data within errors. The associated polarization fractions of those vector-vector and vector-tensor decay modes are also predicted, which are expected to be tested in the near future experiments. The invariant mass spectra for the corresponding resonances in the $B_{(s)} \to [D^{(*)}, \bar{D}^{(*)} ] K^+K^-$ decays are well established, which can be confronted with the precise data from the LHCb and Belle II experiments.

hep-ph↗
arXiv2020

$P$-wave contributions to $B_{(s)}\toψKπ$ decays in perturbative QCD approach

In this work, we studied the quasi-two-body decays $B_{(s)} \to ψ[K^*(892), K^*(1410),$ $K^*(1680)] \to ψKπ$ by employing the perturbative QCD (PQCD) factorization approach, where the charmonia $ψ$ represents $J/ψ$ and $ψ(2S)$. The corresponding decay channels are studied by constructing the kaon-pion distribution amplitude (DA) $Φ_{K π}^{\rm P}$, which contained the important final state interactions between the kaon and pion in the resonant region. The relativistic Breit-Wigner formulas are adopted to parameterize the time-like form factor $F_{Kπ}$ appeared in the kaon-pion DAs. The SU(3) flavor symmetry breaking effect resulting from the mass difference between kaon and pion is taken into account, which makes significant contributions to the longitudinal polarizations. We accommodate well the observed branching ratios and the polarization fractions of the $B_{(s)} \to ψK^*(892) \to ψKπ$ by tuning the hadronic parameters for the kaon-pion DAs. The PQCD predictions for the $B_{(s)} \to ψ[K^*(1410), K^*(1680)] \to ψKπ$ modes from the same set of parameters can be tested by the future precise data from the LHCb and the Belle II experiments.

hep-ph↗
arXiv2020

$S$, $P$ and $D$-wave resonance contributions to $B_{(s)} \to η_c(1S,2S) Kπ$ decays in the perturbative QCD approach

In this work, we analyze the three-body $B_{(s)} \to η_c(1S,2S) K π$ decays within the framework of the perturbative QCD approach (PQCD) under the quasi-two-body approximation, where the kaon-pion invariant mass spectra are dominated by the $K_0^*(1430)^0,K_0^*(1950)^0,K^*(892)^0,K^*(1410)^0,K^*(1680)^0$ and $K_2^*(1430)^0$ resonances. The time-like form factors are adopted to parametrize the corresponding $S$, $P$, $D$-wave kaon-pion distribution amplitudes for the concerned decay modes, which describe the final-state interactions between the kaon and pion in the resonant region. The $Kπ$ $S$-wave component at low $Kπ$ mass is described by the LASS line shape, while the time-like form factors of other resonances are modeled by the relativistic Breit-Wigner function. We find the following main points: (a) the PQCD predictions of the branching ratios for most considered $B \to η_c(1S)(K^{*0}\to )K^+π^-$ decays agree well with the currently available data within errors; (b) for ${\cal B}(B^0 \to η_c (K_0^*(1430)\to )K^+π^-)$ and ${\cal B}(B^0 \to η_c K^+π^-({\rm NR}))$ (here NR means nonresonant), our predictions of the branching ratios are a bit smaller than the measured ones; and (c) the PQCD results for the $D$-wave contributions considered in this work can be tested once the precise data from the future LHCb and Belle-II experiments are available.

hep-ph↗
arXiv2020

Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking

The success of DNNs has driven the extensive applications of person re-identification (ReID) into a new era. However, whether ReID inherits the vulnerability of DNNs remains unexplored. To examine the robustness of ReID systems is rather important because the insecurity of ReID systems may cause severe losses, e.g., the criminals may use the adversarial perturbations to cheat the CCTV systems. In this work, we examine the insecurity of current best-performing ReID models by proposing a learning-to-mis-rank formulation to perturb the ranking of the system output. As the cross-dataset transferability is crucial in the ReID domain, we also perform a back-box attack by developing a novel multi-stage network architecture that pyramids the features of different levels to extract general and transferable features for the adversarial perturbations. Our method can control the number of malicious pixels by using differentiable multi-shot sampling. To guarantee the inconspicuousness of the attack, we also propose a new perception loss to achieve better visual quality. Extensive experiments on four of the largest ReID benchmarks (i.e., Market1501 [45], CUHK03 [18], DukeMTMC [33], and MSMT17 [40]) not only show the effectiveness of our method, but also provides directions of the future improvement in the robustness of ReID systems. For example, the accuracy of one of the best-performing ReID systems drops sharply from 91.8% to 1.4% after being attacked by our method. Some attack results are shown in Fig. 1. The code is available at https://github.com/whj363636/Adversarial-attack-on-Person-ReID-With-Deep-Mis-Ranking.

cs.CV↗
arXiv2019

Expression Analysis Based on Face Regions in Read-world Conditions

Facial emotion recognition is an essential and important aspect of the field of human-machine interaction. Past research on facial emotion recognition focuses on the laboratory environment. However, it faces many challenges in real-world conditions, i.e., illumination changes, large pose variations and partial or full occlusions. Those challenges lead to different face areas with different degrees of sharpness and completeness. Inspired by this fact, we focus on the authenticity of predictions generated by different pairs. For example, if only the mouth areas are available and the emotion classifier predicts happiness, then there is a question of how to judge the authenticity of predictions. This problem can be converted into the contribution of different face areas to different emotions. In this paper, we divide the whole face into six areas: nose areas, mouth areas, eyes areas, nose to mouth areas, nose to eyes areas and mouth to eyes areas. To obtain more convincing results, our experiments are conducted on three different databases: facial expression recognition + ( FER+), real-world affective faces database (RAF-DB) and expression in-the-wild (ExpW) dataset. Through analysis of the classification accuracy, the confusion matrix and the class activation map (CAM), we can establish convincing results. To sum up, the contributions of this paper lie in two areas: 1) We visualize concerned areas of human faces in emotion recognition; 2) We analyze the contribution of different face areas to different emotions in real-world conditions through experimental analysis. Our findings can be combined with findings in psychology to promote the understanding of emotional expressions.

cs.CV↗
arXiv2019

Speech Emotion Recognition via Contrastive Loss under Siamese Networks

Speech emotion recognition is an important aspect of human-computer interaction. Prior work proposes various end-to-end models to improve the classification performance. However, most of them rely on the cross-entropy loss together with softmax as the supervision component, which does not explicitly encourage discriminative learning of features. In this paper, we introduce the contrastive loss function to encourage intra-class compactness and inter-class separability between learnable features. Furthermore, multiple feature selection methods and pairwise sample selection methods are evaluated. To verify the performance of the proposed system, we conduct experiments on The Interactive Emotional Dyadic Motion Capture (IEMOCAP) database, a common evaluation corpus. Experimental results reveal the advantages of the proposed method, which reaches 62.19% in the weighted accuracy and 63.21% in the unweighted accuracy. It outperforms the baseline system that is optimized without the contrastive loss function with 1.14% and 2.55% in the weighted accuracy and the unweighted accuracy, respectively.

cs.CV↗
arXiv2019

Studies of the resonance components in the $B_s$ decays into charmonia plus kaon pair

In this work, the decays of $B_s$ mesons to a charmonium state and a $K^+K^-$ pair are carefully investigated in the perturbative QCD approach. Following the latest fit from the LHCb experiment, we restrict ourselves to the case where the produced $K^+K^-$ pair interact in isospin zero $S$, $P$, and $D$ wave resonances in the kinematically allowed mass window. Besides the dominant contributions of the $ϕ(1020)$ resonance in the $P$-wave and $f_2'(1525)$ in the $D$-wave, other resonant structures in the high mass region as well as the $S$-wave components are also included. The invariant mass spectra for most of the resonances in the $B_s\rightarrow J/ψK^+K^-$ decay are well reproduced. The obtained three-body decay branching ratios can reach the order of $10^{-4}$, which seem to be accessible in the near future experiments. The associated polarization fractions of those vector-vector and vector-tensor modes are also predicted, which are compared with the existing data from LHCb Collaboration.

hep-ph↗
arXiv2019

Quasi-two-body decays $B \to D K^*(892) \to D K π$ in the perturbative QCD approach

We study the quasi-two-body decays $B\to D K^*(892) \to D Kπ$ by employing the perturbative QCD approach. The two-meson distribution amplitudes $Φ_{Kπ}^{\text{$P$-wave}}$ are adopted to describe the final state interactions of the kaon-pion pair in the resonance region. The resonance line shape for the $P$-wave $Kπ$ component $K^*(892)$ in the time-like form factor $F_{Kπ}(s)$ is parameterized by the relativistic Breit-Wigner function. For most considered decay modes, the theoretical predictions for their branching ratios are consistent with currently available experimental measurements within errors. We also disscuss some ratios of the branching fractions of the concerned decay processes. More precise data from LHCb and Belle-II are expected to test our predictions.

hep-ph↗