arXiv subjects

Yuliya Dovzhenko

Publications and source records attributed to Yuliya Dovzhenko.

3 recordsLinked to original sources

arXiv2021

Imaging phonon-mediated hydrodynamic flow in WTe2

In the presence of interactions, electrons in condensed-matter systems can behave hydrodynamically, exhibiting phenomena associated with classical fluids, such as vortices and Poiseuille flow. In most conductors, electron-electron interactions are minimized by screening effects, hindering the search for hydrodynamic materials; however, recently, a class of semimetals has been reported to exhibit prominent interactions. Here we study the current flow in the layered semimetal tungsten ditelluride by imaging the local magnetic field using a nitrogen-vacancy defect in a diamond. We image the spatial current profile within three-dimensional tungsten ditelluride and find that it exhibits non-uniform current density, indicating hydrodynamic flow. Our temperature-resolve current profile measurements reveal a non-monotonic temperature dependence, with the strongest hydrodynamic effects at approximately 20 K. We also report ab initio calculations showing that electron-electron interactions are not explained by the Coulomb interaction alone, but are predominantly mediated by phonons. This provides a promising avenue in the search for hydrodynamic flow and prominent electron interactions in high-carrier-density materials.

cond-mat.mes-hall↗

arXiv2021

SPGISpeech: 5,000 hours of transcribed financial audio for fully formatted end-to-end speech recognition

In the English speech-to-text (STT) machine learning task, acoustic models are conventionally trained on uncased Latin characters, and any necessary orthography (such as capitalization, punctuation, and denormalization of non-standard words) is imputed by separate post-processing models. This adds complexity and limits performance, as many formatting tasks benefit from semantic information present in the acoustic signal but absent in transcription. Here we propose a new STT task: end-to-end neural transcription with fully formatted text for target labels. We present baseline Conformer-based models trained on a corpus of 5,000 hours of professionally transcribed earnings calls, achieving a CER of 1.7. As a contribution to the STT research community, we release the corpus free for non-commercial use at https://datasets.kensho.com/datasets/scribe.

cs.CL↗

arXiv2017

Imaging the Spin Texture of a Skyrmion Under Ambient Conditions Using an Atomic-Sized Sensor

Magnetic skyrmions are two-dimensional non-collinear spin textures characterised by an integer topological number. They commonly crystallise at low temperatures in bulk noncentrosymmetric ferromagnets where the lack of inversion symmetry leads to an antisymmetric component of the exchange interaction. Recently, stable room-temperature skyrmions were reported in stacks of thin magnetic films where antisymmetric exchange results from broken symmetry at the interface. Determining the spin structure of these technologically-relevant skyrmions in the presence of external magnetic fields has remained a key experimental challenge. Here, we use the single electron spin of a Nitrogen-Vacancy (NV) centre in diamond to perform scanning magnetometry of skyrmions in Pt/Co/Ta multilayers under ambient conditions. We introduce a novel method to assess the manifold of spin textures compatible with the measured local magnetic fields and identify physically allowed configurations based on the topology of the resulting solution. We determine that the underlying magnetization pattern for the skyrmion is consistent with a cycloid (or Néel)-like spin texture, in agreement with theoretical predictions for interfacial antisymmetric exchange interaction. Our results open up wide-ranging possibilities in quantitative model-free imaging of two-dimensional spin structures using scanning NV magnetometry.

cond-mat.str-el↗