77 citations · 243 across the 31 of their papers we have counts for
31 papers
End-to-End Neural Decomposition with Koopman Operators for Time-Series Forecasting
De-Yan Lu, Xugang Lu, Yu Tsao +1
Koopman theory offers a linear-operator view of nonlinear sequence dynamics by lifting observations into a space where evolution is governed by a linear time-invariant Koopman oper…
Multi-Level Knowledge Distillation for Speech Emotion Recognition in Noisy Conditions
Yang Liu, Haoqin Sun, Geng Chen +4
Speech emotion recognition (SER) performance deteriorates significantly in the presence of noise, making it challenging to achieve competitive performance in noisy conditions. To t…
Neural domain alignment for spoken language recognition based on optimal transport
Xugang Lu, Peng Shen, Yu Tsao +1
Domain shift poses a significant challenge in cross-domain spoken language recognition (SLR) by reducing its effectiveness. Unsupervised domain adaptation (UDA) algorithms have bee…
Hierarchical Cross-Modality Knowledge Transfer with Sinkhorn Attention for CTC-based ASR
Xugang Lu, Peng Shen, Yu Tsao +1
Due to the modality discrepancy between textual and acoustic modeling, efficiently transferring linguistic knowledge from a pretrained language model (PLM) to acoustic encoding for…
Cross-modal Alignment with Optimal Transport for CTC-based ASR
Xugang Lu, Peng Shen, Yu Tsao +1
Temporal connectionist temporal classification (CTC)-based automatic speech recognition (ASR) is one of the most successful end to end (E2E) ASR frameworks. However, due to the tok…
Pronunciation-aware unique character encoding for RNN Transducer-based Mandarin speech recognition
Peng Shen, Xugang Lu, Hisashi Kawai
For Mandarin end-to-end (E2E) automatic speech recognition (ASR) tasks, compared to character-based modeling units, pronunciation-based modeling units could improve the sharing of…
Transducer-based language embedding for spoken language identification
Peng Shen, Xugang Lu, Hisashi Kawai
The acoustic and linguistic features are important cues for the spoken language identification (LID) task. Recent advanced LID systems mainly use acoustic features that lack the us…
Partial Coupling of Optimal Transport for Spoken Language Identification
Xugang Lu, Peng Shen, Yu Tsao +1
In order to reduce domain discrepancy to improve the performance of cross-domain spoken language identification (SLID) system, as an unsupervised domain adaptation (UDA) method, we…
TMS: A Temporal Multi-scale Backbone Design for Speaker Embedding
Ruiteng Zhang, Jianguo Wei, Xugang Lu +6
Speaker embedding is an important front-end module to explore discriminative speaker features for many speech applications where speaker information is needed. Current SOTA backbon…
Perceptual Contrast Stretching on Target Feature for Speech Enhancement
Rong Chao, Cheng Yu, Szu-Wei Fu +2
Speech enhancement (SE) performance has improved considerably owing to the use of deep learning models as a base function. Herein, we propose a perceptual contrast stretching (PCS)…
A Novel Temporal Attentive-Pooling based Convolutional Recurrent Architecture for Acoustic Signal Enhancement
Tassadaq Hussain, Wei-Chien Wang, Mandar Gogate +5
In acoustic signal processing, the target signals usually carry semantic information, which is encoded in a hierarchal structure of short and long-term contexts. However, the backg…
Unsupervised Noise Adaptive Speech Enhancement by Discriminator-Constrained Optimal Transport
Hsin-Yi Lin, Huan-Hsin Tseng, Xugang Lu +1
This paper presents a novel discriminator-constrained optimal transport network (DOTN) that performs unsupervised domain adaptation for speech enhancement (SE), which is an essenti…
CS-Rep: Making Speaker Verification Networks Embracing Re-parameterization
Ruiteng Zhang, Jianguo Wei, Wenhuan Lu +4
Automatic speaker verification (ASV) systems, which determine whether two speeches are from the same speaker, mainly focus on verification accuracy while ignoring inference speed.…
Siamese Neural Network with Joint Bayesian Model Structure for Speaker Verification
Xugang Lu, Peng Shen, Yu Tsao +1
Generative probability models are widely used for speaker verification (SV). However, the generative models are lack of discriminative feature selection ability. As a hypothesis te…
MetricGAN+: An Improved Version of MetricGAN for Speech Enhancement
Szu-Wei Fu, Cheng Yu, Tsun-An Hsieh +4
The discrepancy between the cost function used for training a speech enhancement model and human auditory perception usually makes the quality of enhanced speech unsatisfactory. Ob…
EMA2S: An End-to-End Multimodal Articulatory-to-Speech System
Yu-Wen Chen, Kuo-Hsuan Hung, Shang-Yi Chuang +4
Synthesized speech from articulatory movements can have real-world use for patients with vocal cord disorders, situations requiring silent speech, or in high-noise environments. In…
Coupling a generative model with a discriminative learning framework for speaker verification
Xugang Lu, Peng Shen, Yu Tsao +1
The speaker verification (SV) task is to decide whether an utterance is spoken by a target or an imposter speaker. For most studies, a log-likelihood ratio (LLR) score is estimated…
Unsupervised neural adaptation model based on optimal transport for spoken language identification
Xugang Lu, Peng Shen, Yu Tsao +1
Due to the mismatch of statistical distributions of acoustic speech between training and testing sets, the performance of spoken language identification (SLID) could be drastically…
A Study of Incorporating Articulatory Movement Information in Speech Enhancement
Yu-Wen Chen, Kuo-Hsuan Hung, Shang-Yi Chuang +3
Although deep learning algorithms are widely used for improving speech enhancement (SE) performance, the performance remains limited under highly challenging conditions, such as un…
Improving Perceptual Quality by Phone-Fortified Perceptual Loss using Wasserstein Distance for Speech Enhancement
Tsun-An Hsieh, Cheng Yu, Szu-Wei Fu +2
Speech enhancement (SE) aims to improve speech quality and intelligibility, which are both related to a smooth transition in speech segments that may carry linguistic information,…
Incorporating Broad Phonetic Information for Speech Enhancement
Yen-Ju Lu, Chien-Feng Liao, Xugang Lu +2
In noisy conditions, knowing speech contents facilitates listeners to more effectively suppress background noise components and to retrieve pure speech signals. Previous studies ha…
WaveCRN: An Efficient Convolutional Recurrent Neural Network for End-to-end Speech Enhancement
Tsun-An Hsieh, Hsin-Min Wang, Xugang Lu +1
Due to the simple design pipeline, end-to-end (E2E) neural models for speech enhancement (SE) have attracted great interest. In order to improve the performance of the E2E model, t…
Speech Enhancement based on Denoising Autoencoder with Multi-branched Encoders
Cheng Yu, Ryandhimas E. Zezario, Syu-Siang Wang +5
Deep learning-based models have greatly advanced the performance of speech enhancement (SE) systems. However, two problems remain unsolved, which are closely related to model gener…
Cross-scale Attention Model for Acoustic Event Classification
Xugang Lu, Peng Shen, Sheng Li +2
A major advantage of a deep convolutional neural network (CNN) is that the focused receptive field size is increased by stacking multiple convolutional layers. Accordingly, the mod…