arXiv subjects

Siyang Wang

Publications and source records attributed to Siyang Wang.

At least 37 records · Page 2Linked to original sources

arXiv2024

Upwind summation-by-parts finite differences: error estimates and WENO methodology

High order upwind summation-by-parts finite difference operators have recently been developed. When combined with the simultaneous-approximation-term method to impose boundary conditions, the method converges faster than using traditional summation-by-parts operators. We prove the convergence rate by the normal mode analysis for such methods for a class of hyperbolic partial differential equations. Our analysis shows that the penalty parameter for imposing boundary conditions affects the convergence rate for stable methods. In addition, to solve problems with discontinuous data, we extend the method to also have the weighted essentially nonoscillatory property. The overall method is stable, achieves high order accuracy for smooth problems, and is capable of solving problems with discontinuities.

math.NA↗

arXiv2024

Evaluating Text-to-Speech Synthesis from a Large Discrete Token-based Speech Language Model

Recent advances in generative language modeling applied to discrete speech tokens presented a new avenue for text-to-speech (TTS) synthesis. These speech language models (SLMs), similarly to their textual counterparts, are scalable, probabilistic, and context-aware. While they can produce diverse and natural outputs, they sometimes face issues such as unintelligibility and the inclusion of non-speech noises or hallucination. As the adoption of this innovative paradigm in speech synthesis increases, there is a clear need for an in-depth evaluation of its capabilities and limitations. In this paper, we evaluate TTS from a discrete token-based SLM, through both automatic metrics and listening tests. We examine five key dimensions: speaking style, intelligibility, speaker consistency, prosodic variation, spontaneous behaviour. Our results highlight the model's strength in generating varied prosody and spontaneous outputs. It is also rated higher in naturalness and context appropriateness in listening tests compared to a conventional TTS. However, the model's performance in intelligibility and speaker consistency lags behind traditional TTS. Additionally, we show that increasing the scale of SLMs offers a modest boost in robustness. Our findings aim to serve as a benchmark for future advancements in generative SLMs for speech synthesis.

eess.AS↗

arXiv2023

Shear and delamination behaviour of basal planes in Zr3AlC2 MAX phase studied by micromechanical testing

The mechanical properties of layered, hexagonal-structured MAX phases often show the combined merits of metals and ceramics, making them promising material candidates for safety critical applications. While their unique mechanical performance largely arises from the crystal structure, the effect of chemistry on the properties of these materials remains unclear. To study this, here we employed two in situ electron microscope small scale testing approaches to examine the micromechanical properties of Zr3AlC2, and compared the results with the properties of Ti3SiC2: we used micropillar compression tests to measure basal slip strength, and double cantilever beam splitting tests to evaluate fracture energy for basal plane delamination. We observed distinct and systematic differences in these measured properties between Zr3AlC2 and Ti3SiC2, where Zr3AlC2 appeared to be stronger but more brittle at the microscale, and discussed the implications of the results in the selection, design, and engineering of MAX phases for targeted engineering applications.

cond-mat.mtrl-sci↗

arXiv2023

Determining the Fundamental Failure Modes in Ni-rich Lithium Ion Battery Cathodes

Challenges associated with in-service mechanical degradation of Li-ion battery cathodes has prompted a transition from polycrystalline to single crystal cathode materials. Whilst for single crystal materials, dislocation-assisted crack formation is assumed to be the dominating failure mechanism throughout battery life, there is little direct information about their mechanical behaviour, and mechanistic understanding remains elusive. Here, we demonstrated, using in situ micromechanical testing, direct measurement of local mechanical properties within LiNi0.8Mn0.1Co0.1O2 single crystalline domains. We elucidated the dislocation slip systems, their critical stresses, and how slip facilitate cracking. We then compared single crystal and polycrystal deformation behaviour. Our findings answer two fundamental questions critical to understanding cathode degradation: What dislocation slip systems operate in Ni-rich cathode materials? And how does slip cause fracture? This knowledge unlocks our ability to develop tools for lifetime prediction and failure risk assessment, as well as in designing novel cathode materials with increased toughness in-service.

cond-mat.mtrl-sci↗

arXiv2023

Diff-TTSG: Denoising probabilistic integrated speech and gesture synthesis

With read-aloud speech synthesis achieving high naturalness scores, there is a growing research interest in synthesising spontaneous speech. However, human spontaneous face-to-face conversation has both spoken and non-verbal aspects (here, co-speech gestures). Only recently has research begun to explore the benefits of jointly synthesising these two modalities in a single system. The previous state of the art used non-probabilistic methods, which fail to capture the variability of human speech and motion, and risk producing oversmoothing artefacts and sub-optimal synthesis quality. We present the first diffusion-based probabilistic model, called Diff-TTSG, that jointly learns to synthesise speech and gestures together. Our method can be trained on small datasets from scratch. Furthermore, we describe a set of careful uni- and multi-modal subjective tests for evaluating integrated speech and gesture synthesis systems, and use them to validate our proposed approach. Please see https://shivammehta25.github.io/Diff-TTSG/ for video examples, data, and code.

eess.AS↗

arXiv2023

On the Use of Self-Supervised Speech Representations in Spontaneous Speech Synthesis

Self-supervised learning (SSL) speech representations learned from large amounts of diverse, mixed-quality speech data without transcriptions are gaining ground in many speech technology applications. Prior work has shown that SSL is an effective intermediate representation in two-stage text-to-speech (TTS) for both read and spontaneous speech. However, it is still not clear which SSL and which layer from each SSL model is most suited for spontaneous TTS. We address this shortcoming by extending the scope of comparison for SSL in spontaneous TTS to 6 different SSLs and 3 layers within each SSL. Furthermore, SSL has also shown potential in predicting the mean opinion scores (MOS) of synthesized speech, but this has only been done in read-speech MOS prediction. We extend an SSL-based MOS prediction framework previously developed for scoring read speech synthesis and evaluate its performance on synthesized spontaneous speech. All experiments are conducted twice on two different spontaneous corpora in order to find generalizable trends. Overall, we present comprehensive experimental results on the use of SSL in spontaneous TTS and MOS prediction to further quantify and understand how SSL can be used in spontaneous TTS. Audios samples: https://www.speech.kth.se/tts-demos/sp_ssl_tts

eess.AS↗

arXiv2023

A Comparative Study of Self-Supervised Speech Representations in Read and Spontaneous TTS

Recent work has explored using self-supervised learning (SSL) speech representations such as wav2vec2.0 as the representation medium in standard two-stage TTS, in place of conventionally used mel-spectrograms. It is however unclear which speech SSL is the better fit for TTS, and whether or not the performance differs between read and spontaneous TTS, the later of which is arguably more challenging. This study aims at addressing these questions by testing several speech SSLs, including different layers of the same SSL, in two-stage TTS on both read and spontaneous corpora, while maintaining constant TTS model architecture and training settings. Results from listening tests show that the 9th layer of 12-layer wav2vec2.0 (ASR finetuned) outperforms other tested SSLs and mel-spectrogram, in both read and spontaneous TTS. Our work sheds light on both how speech SSL can readily improve current TTS systems, and how SSLs compare in the challenging generative task of TTS. Audio examples can be found at https://www.speech.kth.se/tts-demos/ssr_tts

eess.AS↗

arXiv2023

High-performance ultrafast pulse compression in the visible spectral range for extreme nonlinear optics at kHz-MHz repetition rates

We demonstrate a remarkably effective single-stage compression technique for ultrafast pulses in the visible electromagnetic spectrum using second-harmonic pulses at 515 nmderived from a 1030 nm Yb-based femtosecond regenerative amplifier. By employing an advanced multi-plate scheme, we achieve more than fourfold compression from 180 fs to 40 fs with an extremely high spectral broadening efficiency of over 95%, and a temporal compression efficiency exceeding 75%. In addition, our method leverages a low nonlinearity medium to attain the shortest pulse durations for a single compressor while maintaining a superb spatial beam quality with 97% of the energy confined in the main lobe of the Arie disk. Moreover, our technique enhances the temporal pulse quality at 515 nm without generating substantial femtosecond-to-picosecond pulse pedestals. The resulting intense visible laser pulses with excellent spatio-temporal parameters and high repetition rate of 100 kHz to 1 MHz open up new frontiers for extreme nonlinear optics and ultrabright EUV and X-ray high-harmonic generation using short VIS wavelength.

physics.optics↗

arXiv2023

Continuously Red-Shift and Blue-Shift Wavelength-Tuneable, Narrowband, High Harmonics in the EUV - X-ray Regime for Resonance Imaging and Spectroscopies

We demonstrate a novel technique for producing high-order harmonics with designer spectral combs in the extreme ultraviolet-soft X-ray range for resonance applications using spectrally controlled visible lasers. Our approach enables continuous tunability of the harmonic peaks while maintaining superb laser-like features such as coherence, narrow bandwidth, and brightness. The harmonics are conveniently shifted towards lower or higher energies by varying the infrared pulse parameters, second harmonic generation phase-matching conditions, and gas density inside a spectral-broadening waveguide. In the time domain, the X-rays are estimated to emerge as a train of sub-300 attosecond pulses, making this source ideal for studying dynamic processes in ferromagnetic nanostructures and other materials through resonant multidimensional coherent diffractive imaging or other X-ray absorption spectroscopy techniques. Moreover, the visible driving laser beams exhibit an ultrashort sub-10 fs pulse dues to nonlinear self-compression with a more than 30-fold enhancement in peak intensity that also extends the tunability of the linewidth of the harmonic combs.

physics.optics↗

arXiv2023

Ultrafast Ultraviolet C and Visible Laser Light with Broadly Tunable Spectrum

We demonstrate a versatile technique for generating continuously wavelength-tunable laser waveforms, with mJ pulse energies and ultrashort pulse durations down to few-cycle in the ultraviolet C and visible spectral ranges. Using the processes of self-phase modulation or Raman-induced spectral broadening, we substantially expand the spectrum of a femtosecond 1030 nm Yb:CaF$_{2}$ laser, allowing for an extensive wavelength-tunability of the second and fourth harmonics of the laser within the visible and ultraviolet C spectral regions at 460-580 nm and 230-290 nm. In addition, our approach exploits nonlinearly assisted self-compression in the second harmonic upconversion of the spectrally broadened infrared pulses with high-order dispersion. This results in 8-30 femtosecond pulses in the visible with an intensity enhancement of up to 30 times. Such an ultrashort visible and ultraviolet C source is ideal for investigating ultrafast dynamics in molecules, solids, and bio and nano systems. Furthermore, this tunable light source enables the generation of bright, narrow-bandwidth, continuously wavelength-tunable, coherent light in the extreme ultraviolet to soft X-ray spectral region. Theoretically, the X-ray pulse structure can consist of sub-200 as pulse trains, making such a source highly suitable for dynamic multidimensional imaging. This includes coherent diffractive imaging of ferromagnetic nanostructures where resonant X-ray scattering is essential, as well as X-ray absorption spectroscopies of advanced quantum materials at pico-nanometer spatial and atto-femtosecond temporal scales.

physics.optics↗

arXiv2023

Automatic Evaluation of Turn-taking Cues in Conversational Speech Synthesis

Turn-taking is a fundamental aspect of human communication where speakers convey their intention to either hold, or yield, their turn through prosodic cues. Using the recently proposed Voice Activity Projection model, we propose an automatic evaluation approach to measure these aspects for conversational speech synthesis. We investigate the ability of three commercial, and two open-source, Text-To-Speech (TTS) systems ability to generate turn-taking cues over simulated turns. By varying the stimuli, or controlling the prosody, we analyze the models performances. We show that while commercial TTS largely provide appropriate cues, they often produce ambiguous signals, and that further improvements are possible. TTS, trained on read or spontaneous speech, produce strong turn-hold but weak turn-yield cues. We argue that this approach, that focus on functional aspects of interaction, provides a useful addition to other important speech metrics, such as intelligibility and naturalness.

eess.AS↗

arXiv2023

Fracture properties of La(Fe,Mn,Si)13 magnetocaloric materials

La(Fe,Mn,Si)13 alloys are a promising material family for magnetic refrigeration. Challenges associated with their structural integrity during device assembly and operation requires deep understanding of the mechanical properties. Here we developed a workflow to quantitatively study the fracture properties of La(Fe,Mn,Si)13 plates used in magnetic cooling devices. We employed microstructural characterisation, optical examination of defects, and four-point bending tests of samples with known defect sizes to evaluate their mechanical performance. We established the residual strength curve which directly links observed defects to mechanical strength. The estimated fracture toughness KC of hydrogenated La(Fe,Mn,Si)13 is approximately 4 MPa m^1/2 for the geometry employed. The established relationship between strength and crack length enables the prediction of mechanical performance through examination of defects via optical microscopy, therefore can be used industrially for directing plate selection to guarantee the mechanical stability of refrigeration devices.

cond-mat.mtrl-sci↗

arXiv2022

A finite difference - discontinuous Galerkin method for the wave equation in second order form

We develop a hybrid spatial discretization for the wave equation in second order form, based on high-order accurate finite difference methods and discontinuous Galerkin methods. The hybridization combines computational efficiency of finite difference methods on Cartesian grids and geometrical flexibility of discontinuous Galerkin methods on unstructured meshes. The two spatial discretizations are coupled by a penalty technique at the interface such that the overall semidiscretization satisfies a discrete energy estimate to ensure stability. In addition, optimal convergence is obtained in the sense that when combining a fourth order finite difference method with a discontinuous Galerkin method using third order local polynomials, the overall convergence rate is fourth order. Furthermore, we use a novel approach to derive an error estimate for the semidiscretization by combining the energy method and the normal mode analysis for a corresponding one dimensional model problem. The stability and accuracy analysis are verified in numerical experiments.

math.NA↗

arXiv2022

An energy-based summation-by-parts finite difference method for the wave equation in second order form

We develop a new finite difference method for the wave equation in second order form. The finite difference operators satisfy a summation-by-parts (SBP) property. With boundary conditions and material interface conditions imposed weakly by the simultaneous-approximation-term (SAT) method, we derive energy estimates for the semi-discretization. In addition, error estimates are derived by the normal mode analysis. The proposed method is termed as energy-based because of its similarity with the energy-based discontinuous Galerkin method. When imposing the Dirichlet boundary condition and material interface conditions, the traditional SBP-SAT discretization uses a penalty term with a mesh-dependent parameter, which is not needed in our method. Furthermore, numerical dissipation can be added to the discretization through the boundary and interface conditions. We present numerical experiments that verify convergence and robustness of the proposed method.

math.NA↗

arXiv2022

The effect of hydrogen on the multiscale mechanical behaviour of a La(Fe,Mn,Si)13-based magnetocaloric material

Magnetocaloric cooling offers the potential to improve the efficiency of refrigeration devices and hence cut the significant CO2 emissions associated with cooling processes. A critical issue in deployment of this technology is the mechanical degradation of the magnetocaloric material during processing and operation, leading to limited service-life. The mechanical properties of hydrogenated La(Fe,Mn,Si)13-based magnetocaloric material are studied using macroscale bending tests of polycrystalline specimens and in situ micropillar compression tests of single crystal specimens. The impact of hydrogenation on the mechanical properties are quantified. Understanding of the deformation/failure mechanisms is aided by characterization with transmission electron microscopy and atom probe tomography to reveal the arrangement of hydrogen atoms in the crystal lattice. Results indicate that the intrinsic strength of this material is ~3-6 GPa and is dependent on the crystal orientation. Single crystals under compressive load exhibit shearing along specific crystallographic planes. Hydrogen deteriorates the strength of La(Fe,Mn,Si)13 through promotion of transgranular fracture. The weakening effect of hydrogen on single crystals is anisotropic; it is significant upon shearing parallel to the {111} crystallographic planes but is negligible when the shear plane is {001}-oriented. APT analysis suggests that this is associated with the close arrangement of hydrogen atoms on {222} planes.

cond-mat.mtrl-sci↗

arXiv2022

A high order finite difference method for the elastic wave equation in bounded domains with nonconforming interfaces

We develop a stable finite difference method for the elastic wave equation in bounded media, where the material properties can be discontinuous at curved interfaces. The governing equation is discretized in second order form by a fourth or sixth order accurate summation-by-parts operator. The mesh size is determined by the velocity structure of the material, resulting in nonconforming grid interfaces with hanging nodes. We use order-preserving interpolation and the ghost point technique to couple adjacent mesh blocks in an energy-conserving manner, which is supported by a fully discrete stability analysis. In our previous work for the wave equation, two pairs of order-preserving interpolation operators are needed when imposing the interface conditions weakly by a penalty technique. Here, we only use one pair in the ghost point method. In numerical experiments, we demonstrate that the convergence rate is optimal, and is the same as when a globally uniform mesh is used in a single domain. In addition, with a predictor-corrector time integration method, we obtain time stepping stability with stepsize almost the same as given by the usual Courant-Friedrichs-Lewy condition.

math.NA↗

arXiv2021

Environment-assisted crack nucleation in La(Fe,Mn,Si)13-based magnetocaloric materials

Cracking in La(Fe,Si)13-based magnetocaloric materials has been observed to predominantly form around La-rich (La2O3) particles and pose a threat to their long-term structural integrity. To understand the formation of these cracks, FIB cross-sectional polishing followed by SEM characterisation was employed to study local microstructural evolution after air exposure. Results suggest that volume expansion and internal degradation associated with a chemical reaction between La2O3 particles and water/moisture can lead to crack nucleation in the 1:13 phase adjacent to La-rich particles. This observation indicates that the formation of La-rich phase should be suppressed, or their size minimised during material processing to ensure the long-term structural integrity of La(Fe,Mn,Si)13 magnetocaloric materials.

cond-mat.mtrl-sci↗

arXiv2021

Nanoscale characterisation of hydrides and secondary phase particles in Zircaloy-4

The interaction of hydrogen and metals continues to be industrially relevant and is a critical part of creating and supporting a safety case for nuclear reactor operation. In the present work, we explore hydrogen storage and hydride formation in a zirconium alloy. We characterise the structure and interfaces of fine scale hydrides using scanning transmission electron microscopy (STEM) including energy dispersive spectroscopy (EDS/EDX), electron energy loss spectroscopy (EELS), and high-resolution STEM. Chemical characterisation is supported further with atom probe tomography (APT). Samples were prepared with cryo-focussed ion beam machining (cryo-FIB) and contain hydrides in α-Zr matrix and hydrides associated with one FeCrZr secondary phase particle (SPP). Major findings include characterisation of different interface planes based upon the size of the hydrides and chemical redistribution of solute ahead of the hydride-metal interface. We also find significant (up to 6 at%) hydrogen retained in solution within the zirconium matrix and show a hydride with only 17 at% hydrogen, which is well below that of a ξ-phase stoichiometry suggesting it is an embryonic hydride. These findings help us understand the distribution of hydrogen and the nanoscale morphology of hydrides, which may influence the lifetime of zirconium-based nuclear fuel cladding.

cond-mat.mtrl-sci↗