Source author record

Yikang Wang

Yikang Wang appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

astro-ph.SR eess.AS Sound astro-ph.GA physics.space-ph

Catalog footprint

What is connected

5works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Combination of Time-domain, Frequency-domain, and Cepstral-domain Acoustic Features for Speech Commands Classification

In speech-related classification tasks, frequency-domain acoustic features such as logarithmic Mel-filter bank coefficients (FBANK) and cepstral-domain acoustic features such as Mel-frequency cepstral coefficients (MFCC) are often used. However, time-domain features perform more effectively in some sound classification tasks which contain non-vocal or weakly speech-related sounds. We previously proposed a feature called bit sequence representation (BSR), which is a time-domain binary acoustic feature based on the raw waveform. Compared with MFCC, BSR performed better in environmental sound detection and showed comparable accuracy performance in limited-vocabulary speech recognition tasks. In this paper, we propose a novel improvement BSR feature called BSR-float16 to represent floating-point values more precisely. We experimentally demonstrated the complementarity among time-domain, frequency-domain, and cepstral-domain features using a dataset called Speech Commands proposed by Google. Therefore, we used a simple back-end score fusion method to improve the final classification accuracy. The fusion results also showed better noise robustness.

preprint2022arXiv

The DKU-OPPO System for the 2022 Spoofing-Aware Speaker Verification Challenge

This paper describes our DKU-OPPO system for the 2022 Spoofing-Aware Speaker Verification (SASV) Challenge. First, we split the joint task into speaker verification (SV) and spoofing countermeasure (CM), these two tasks which are optimized separately. For ASV systems, four state-of-the-art methods are employed. For CM systems, we propose two methods on top of the challenge baseline to further improve the performance, namely Embedding Random Sampling Augmentation (ERSA) and One-Class Confusion Loss(OCCL). Second, we also explore whether SV embedding could help improve CM system performance. We observe a dramatic performance degradation of existing CM systems on the domain-mismatched Voxceleb2 dataset. Third, we compare different fusion strategies, including parallel score fusion and sequential cascaded systems. Compared to the 1.71% SASV-EER baseline, our submitted cascaded system obtains a 0.21% SASV-EER on the challenge official evaluation set.

preprint2020arXiv

Simulation of Alfven wave propagation in magnetic chromosphere with radiative loss: effects of non-linear mode coupling on chromospheric heating

We perform magnetohydrodynamic (MHD) simulations to investigate the propagation of Alfven wave in magnetic chromosphere. We use the 1.5-dimensional expanding flux tube geometry setting and transverse perturbation at the bottom to generate the Alfven wave. Compared with previous studies, our expansion is that we include the radiative loss term introduced by Carlsson & Leenaarts (2012). We find that when an observation based transverse wave generator is applied, the spatial distribution of the time-averaged radiative loss profile in our simulation is consistent with that in the classic atmospheric model. In addition, the energy flux in the corona is larger than the required value for coronal heating in the quiet region. Our study shows that the Alfven wave driven model has the potential to explain chromospheric heating and transport enough energy to the corona simultaneously.

preprint2016arXiv

Distributions of quasar hosts on the galaxy main-sequence plane

The relation between star formation rates and stellar masses, i.e. the galaxy main sequence, is a useful diagnostic of galaxy evolution. We present the distributions relative to the main sequence of 55 optically-selected PG and 12 near-IR-selected 2MASS quasars at z <= 0.5. We estimate the quasar host stellar masses from Hubble Space Telescope or ground-based AO photometry, and the star formation rates through the mid-infrared aromatic features and far-IR photometry. We find that PG quasar hosts more or less follow the main sequence defined by normal star-forming galaxies while 2MASS quasar hosts lie systematically above the main sequence. PG and 2MASS quasars with higher nuclear luminosities seem to have higher specific SFRs (sSFRs), although there is a large scatter. No trends are seen between sSFRs and SMBH masses, Eddington ratios or even morphology types (ellipticals, spirals and mergers). Our results could be placed in an evolutionary scenario with quasars emerging during the transition from ULIRGs/mergers to ellipticals. However, combined with results at higher redshift, they suggest that quasars can be widely triggered in normal galaxies as long as they contain abundant gas and have ongoing star formation.

preprint2015arXiv

Predicting the Arrival Time of Coronal Mass Ejections with the Graduated Cylindrical Shell and Drag Force Model

Accurately predicting the arrival of coronal mass ejections (CMEs) at the Earth based on remote images is of critical significance in the study of space weather. In this paper, we make a statistical study of 21 Earth directed CMEs, exploring in particular the relationship between CME initial speeds and transit times. The initial speed of a CME is obtained by fitting the CME with the Graduated Cylindrical Shell model and is thus free of projection effects. We then use the drag force model to fit results of the transit time versus the initial speed. By adopting different drag regimes, i.e., the viscous, aerodynamics, and hybrid regimes, we get similar results, with the least mean estimation error of the hybrid model of 12.9 hours. CMEs with a propagation angle (the angle between the propagation direction and the Sun-Earth line) larger than its half angular width arrive at the Earth with an angular deviation caused by factors other than the radial solar wind drag. The drag force model cannot be well applied to such events. If we exclude these events in the sample, the prediction accuracy can be improved, i.e., the estimation error reduces to 6.8 hours. This work suggests that it is viable to predict the arrival time of CMEs at the Earth based on the initial parameters with a fairly good accuracy. Thus, it provides a method of space weather forecast of 1--5 days following the occurrence of CMEs.