arXiv subjects

Nguyen Tien Cuong

Publications and source records attributed to Nguyen Tien Cuong.

3 recordsLinked to original sources

arXiv2026

Domain-Aware Mispronunciation Detection and Diagnosis Using Language-Specific Statistical Graphs

Mispronunciation Detection and Diagnosis (MDD) has gained increasing importance in computer-assisted language learning and speech technology in recent years. In this paper, we propose a method for constructing statistical graphs that enable models to learn phoneme confusion patterns represented as directed graphs. Furthermore, we introduce a language-specific strategy to capture systematic pronunciation differences across various native language (L1) backgrounds. The effectiveness of our approach is demonstrated through extensive experiments on the L2-ARCTIC benchmark, where it achieves an F1-score of 59.52%, outperforming several competitive baselines.

cs.CL↗

arXiv2025

Mispronunciation Detection and Diagnosis Without Model Training: A Retrieval-Based Approach

Mispronunciation Detection and Diagnosis (MDD) is crucial for language learning and speech therapy. Unlike conventional methods that require scoring models or training phoneme-level models, we propose a novel training-free framework that leverages retrieval techniques with a pretrained Automatic Speech Recognition model. Our method avoids phoneme-specific modeling or additional task-specific training, while still achieving accurate detection and diagnosis of pronunciation errors. Experiments on the L2-ARCTIC dataset show that our method achieves a superior F1 score of 69.60% while avoiding the complexity of model training.

cs.CL↗

arXiv2021

First-principles study on controlling energy gap of graphene using hybrid armchair-zigzag nanostructures

The electronic and transport properties of hybrid armchair zigzag nanostructures including U-shaped graphene nanoribbons and patterned nanopores structured graphene were studied using combination of density functional theory and non-equilibrium Green function method. The density of state, electron transmission spectra, and molecular orbitals were analyzed. The obtained results show that GNRs junctions tend to open an energy gap when U-shaped structures were formed due to the formation of quasi-bound states at zigzag edges. The size of U shaped structures has enormous influences on the electron transport of the system. We also considered the effect of corner form of the U-shaped GNRs junctions on energy gap opening. It was found that as some carbon atoms are add to the inner corner, the energy gap in U shaped GNRs significantly changed. For patterned nanopores structured graphene, the calculated results show that patterned nanopores enormous influence on electronic and the transport properties though the GNRs junctions, depending on the shape, size, and the number of nanopores. The study suggests that designed tailored graphene systems based on hybrid armchair zigzag nanostructures can be used to control the energy gap of graphene.

cond-mat.mtrl-sci↗