Jiyang Tang

Jiyang Tang

Auto-generated profile, based on publicly available information

Jiyang Tang

Language Technologies Institute Carnegie Mellon University USA

Research Focus & Metrics

Academic Career Length

32Years (1995-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 November 1995 in Journal of Physics B: Atomic, Molecular and Optical Physics
Verified
Journal Article
Open access
PDF file
TKT Kambara
T Kambara
JTJ Z Tang
J Z Tang
YAY Awaya
Y Awaya
BDB D DePaola
B D DePaola
OJO Jagutzki
O Jagutzki
ISI Shimamura
I Shimamura
Using high-resolution recoil-ion momentum spectroscopy we have measured the differential cross sections of single-electron capture and target single-ionization processes for 8.7 MeV O7+-He collisions as functions of scattering angle. A transverse momentum resolution of +or-0.2 au, which corresponds to an angular resolution of about +or-1.5*10-6 rad for the projectile scattering angle, was obtained by intersecting a well collimated O7+ beam with a target of a supersonic He jet from a pre-cooled gas and by measuring the recoil-ion transverse momentum. For the single capture reaction, information on the n-value of the electron final state in O6+ (1snl) is obtained from the longitudinal momentum of the recoil ions. In pure single-electron capture, the dominant contributions to capture were found to be those from the n=4 and higher states, whereas single capture accompanied by the ionization of the second target electron mainly populates n=2 to n=4 states. Furthermore, the measured transverse momentum distribution differs significantly between pure single capture and capture with simultaneous ionization. The measured data for the pure capture process compare favourably with theoretical results based on a molecular-state expansion method. Other experimental data are discussed in terms of the classical overbarrier model.

19 May 2023 in arXiv
Verified
Preprint
Open access
PDF file
JTJiyang Tang
Jiyang Tang
WCWilliam Chen
William Chen
XCXuankai Chang
Xuankai Chang
SWShinji Watanabe
Shinji Watanabe
BMBrian MacWhinney
Brian MacWhinney
Aphasia is a language disorder that affects the speaking ability of millions of patients. This paper presents a new benchmark for Aphasia speech recognition and detection tasks using state-of-the-art speech recognition techniques with the AphsiaBank dataset. Specifically, we introduce two multi-task learning methods based on the CTC/Attention architecture to perform both tasks simultaneously. Our system achieves state-of-the-art speaker-level detection accuracy (97.3%), and a relative WER reduction of 11% for moderate Aphasia patients. In addition, we demonstrate the generalizability of our approach by applying it to another disordered speech database, the DementiaBank Pitt corpus. We will make our all-in-one recipes and pre-trained model publicly available to facilitate reproducibility. Our standardized data preprocessing pipeline and open-source recipes enable researchers to compare results directly, promoting progress in disordered speech processing.