Search

Your search keyword '"Meng Zhong"' showing total 41 results
Start Over Author "Meng Zhong" Publisher ieee
41 results on '"Meng Zhong"'

5. Modular Hybrid Autoregressive Transducer

Author

Meng, Zhong, primary, Chen, Tongzhou, additional, Prabhavalkar, Rohit, additional, Zhang, Yu, additional, Wang, Gary, additional, Audhkhasi, Kartik, additional, Emond, Jesse, additional, Strohman, Trevor, additional, Ramabhadran, Bhuvana, additional, Huang, W. Ronny, additional, Variani, Ehsan, additional, Huang, Yinghui, additional, and Moreno, Pedro J., additional

Published
2023
Full Text
View/download PDF

23. Speech Separation Using Speaker Inventory

Author

Wang, Peidong, primary, Chen, Zhuo, additional, Xiao, Xiong, additional, Meng, Zhong, additional, Yoshioka, Takuya, additional, Zhou, Tianyan, additional, Lu, Liang, additional, and Li, Jinyu, additional

Published
2019
Full Text
View/download PDF

25. Adversarial Teacher-Student Learning for Unsupervised Domain Adaptation

Author

Meng, Zhong, Li, Jinyu, Gong, Yifan, Biing-Hwang, and Juang

Subjects
FOS: Computer and information sciences, Computer Science - Machine Learning, Sound (cs.SD), Computer science, Speech recognition, Feature extraction, Word error rate, 02 engineering and technology, Computer Science - Sound, Machine Learning (cs.LG), Data modeling, 030507 speech-language pathology & audiology, 03 medical and health sciences, Audio and Speech Processing (eess.AS), Robustness (computer science), FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, Student learning, Computer Science - Computation and Language, Acoustic model, 020206 networking & telecommunications, Test set, 0305 other medical science, Transfer of learning, Computation and Language (cs.CL), Classifier (UML), Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

The teacher-student (T/S) learning has been shown effective in unsupervised domain adaptation [1]. It is a form of transfer learning, not in terms of the transfer of recognition decisions, but the knowledge of posteriori probabilities in the source domain as evaluated by the teacher model. It learns to handle the speaker and environment variability inherent in and restricted to the speech signal in the target domain without proactively addressing the robustness to other likely conditions. Performance degradation may thus ensue. In this work, we advance T/S learning by proposing adversarial T/S learning to explicitly achieve condition-robust unsupervised domain adaptation. In this method, a student acoustic model and a condition classifier are jointly optimized to minimize the Kullback-Leibler divergence between the output distributions of the teacher and student models, and simultaneously, to min-maximize the condition classification loss. A condition-invariant deep feature is learned in the adapted student model through this procedure. We further propose multi-factorial adversarial T/S learning which suppresses condition variabilities caused by multiple factors simultaneously. Evaluated with the noisy CHiME-3 test set, the proposed methods achieve relative word error rate improvements of 44.60% and 5.38%, respectively, over a clean source model and a strong T/S learning baseline model., Comment: 5 pages, 1 figure, ICASSP 2018

Published
2018

26. Speaker-Invariant Training Via Adversarial Learning

Author

Meng, Zhong, Li, Jinyu, Chen, Zhuo, Zhao, Yong, Mazalov, Vadim, Gong, Yifan, Biing-Hwang, and Juang

Subjects
FOS: Computer and information sciences, Sound (cs.SD), Computer Science - Artificial Intelligence, Computer science, Speech recognition, Feature extraction, Word error rate, 02 engineering and technology, Computer Science - Sound, 030507 speech-language pathology & audiology, 03 medical and health sciences, Audio and Speech Processing (eess.AS), FOS: Electrical engineering, electronic engineering, information engineering, 0202 electrical engineering, electronic engineering, information engineering, Hidden Markov model, Computer Science - Computation and Language, Noise measurement, Artificial neural network, Acoustic model, 020206 networking & telecommunications, Triphone, Artificial Intelligence (cs.AI), 0305 other medical science, Computation and Language (cs.CL), Decoding methods, Electrical Engineering and Systems Science - Audio and Speech Processing
Abstract

We propose a novel adversarial multi-task learning scheme, aiming at actively curtailing the inter-talker feature variability while maximizing its senone discriminability so as to enhance the performance of a deep neural network (DNN) based ASR system. We call the scheme speaker-invariant training (SIT). In SIT, a DNN acoustic model and a speaker classifier network are jointly optimized to minimize the senone (tied triphone state) classification loss, and simultaneously mini-maximize the speaker classification loss. A speaker-invariant and senone-discriminative deep feature is learned through this adversarial multi-task learning. With SIT, a canonical DNN acoustic model with significantly reduced variance in its output probabilities is learned with no explicit speaker-independent (SI) transformations or speaker-specific representations used in training or testing. Evaluated on the CHiME-3 dataset, the SIT achieves 4.99% relative word error rate (WER) improvement over the conventional SI acoustic model. With additional unsupervised speaker adaptation, the speaker-adapted (SA) SIT model achieves 4.86% relative WER gain over the SA SI acoustic model., 5 pages, 3 figures, ICASSP 2018

Published
2018

34. A novel 3D ray-tracing model for propagation prediction in indoor environments

Author

Zhong-Yu Liu, Lixin Guo, Xiao Meng, and Zhi-Meng Zhong

Subjects
Physics, Ray tracing (physics), Optics, Geometrical optics, business.industry, Uniform theory of diffraction, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Radio channel, Radiowave propagation, Geometrical theory of diffraction, business, ComputingMethodologies_COMPUTERGRAPHICS
Abstract

A novel three-dimensional (3D) ray-tracing model for the use of the uniform theory of diffraction and geometrical optics in radio channel characterizations of indoor environments is presented in this paper. The model is based on a technique which multiple reflections, transmissions and diffractions are considered via the ray-path classification into four different categories. And each ray category is determined by using different methods. At last, our theoretical results of the received power are compared with measurements which have been reported. The good agreement with these measurements indicates that our prediction model works well for such indoor communication applications.

Published
2012

Catalog

Books, media, physical & digital resources
See catalog results