Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Hung-yi Lee

Hung-yi Lee

Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation


May 11, 2021
Shun-Po Chuang, Yung-Sung Chuang, Chih-Chiang Chang, Hung-yi Lee

Figure 1 for Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation
Figure 2 for Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation
Figure 3 for Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation
Figure 4 for Investigating the Reordering Capability in CTC-based Non-Autoregressive End-to-End Speech Translation
* Accepted in Findings of ACL-IJCNLP 2021 
  Access Paper or Ask Questions

Improving Cross-Lingual Reading Comprehension with Self-Training


May 08, 2021
Wei-Cheng Huang, Chien-yu Huang, Hung-yi Lee

Figure 1 for Improving Cross-Lingual Reading Comprehension with Self-Training
Figure 2 for Improving Cross-Lingual Reading Comprehension with Self-Training
Figure 3 for Improving Cross-Lingual Reading Comprehension with Self-Training
Figure 4 for Improving Cross-Lingual Reading Comprehension with Self-Training
* 8 pages, 4 figures 
  Access Paper or Ask Questions

SpeechNet: A Universal Modularized Model for Speech Processing Tasks


May 07, 2021
Yi-Chen Chen, Po-Han Chi, Shu-wen Yang, Kai-Wei Chang, Jheng-hao Lin, Sung-Feng Huang, Da-Rong Liu, Chi-Liang Liu, Cheng-Kuang Lee, Hung-yi Lee

Figure 1 for SpeechNet: A Universal Modularized Model for Speech Processing Tasks
Figure 2 for SpeechNet: A Universal Modularized Model for Speech Processing Tasks
Figure 3 for SpeechNet: A Universal Modularized Model for Speech Processing Tasks
Figure 4 for SpeechNet: A Universal Modularized Model for Speech Processing Tasks

  Access Paper or Ask Questions

SUPERB: Speech processing Universal PERformance Benchmark


May 03, 2021
Shu-wen Yang, Po-Han Chi, Yung-Sung Chuang, Cheng-I Jeff Lai, Kushal Lakhotia, Yist Y. Lin, Andy T. Liu, Jiatong Shi, Xuankai Chang, Guan-Ting Lin, Tzu-Hsien Huang, Wei-Cheng Tseng, Ko-tik Lee, Da-Rong Liu, Zili Huang, Shuyan Dong, Shang-Wen Li, Shinji Watanabe, Abdelrahman Mohamed, Hung-yi Lee

Figure 1 for SUPERB: Speech processing Universal PERformance Benchmark
Figure 2 for SUPERB: Speech processing Universal PERformance Benchmark
* Submitted to Interspeech 2021 
  Access Paper or Ask Questions

Put Chatbot into Its Interlocutor's Shoes: New Framework to Learn Chatbot Responding with Intention


Apr 23, 2021
Hsuan Su, Jiun-Hao Jhan, Fan-yun Sun, Saurav Sahay, Hung-yi Lee

Figure 1 for Put Chatbot into Its Interlocutor's Shoes: New Framework to Learn Chatbot Responding with Intention
Figure 2 for Put Chatbot into Its Interlocutor's Shoes: New Framework to Learn Chatbot Responding with Intention
Figure 3 for Put Chatbot into Its Interlocutor's Shoes: New Framework to Learn Chatbot Responding with Intention
Figure 4 for Put Chatbot into Its Interlocutor's Shoes: New Framework to Learn Chatbot Responding with Intention
* Accepted at NAACL-HLT 2021 
  Access Paper or Ask Questions

Utilizing Self-supervised Representations for MOS Prediction


Apr 21, 2021
Wei-Cheng Tseng, Chien-yu Huang, Wei-Tsung Kao, Yist Y. Lin, Hung-yi Lee

Figure 1 for Utilizing Self-supervised Representations for MOS Prediction
Figure 2 for Utilizing Self-supervised Representations for MOS Prediction
Figure 3 for Utilizing Self-supervised Representations for MOS Prediction
Figure 4 for Utilizing Self-supervised Representations for MOS Prediction
* Submitted to Interspeech 2021. We acknowledge the support of AWS Machine Learning Research Awards program 
  Access Paper or Ask Questions

S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations


Apr 07, 2021
Jheng-hao Lin, Yist Y. Lin, Chung-Ming Chien, Hung-yi Lee

Figure 1 for S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations
Figure 2 for S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations
Figure 3 for S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations
Figure 4 for S2VC: A Framework for Any-to-Any Voice Conversion with Self-Supervised Pretrained Representations
* Submitted to INTERSPEECH 2021 
  Access Paper or Ask Questions

Non-autoregressive Mandarin-English Code-switching Speech Recognition with Pinyin Mask-CTC and Word Embedding Regularization


Apr 06, 2021
Shun-Po Chuang, Heng-Jui Chang, Sung-Feng Huang, Hung-yi Lee

Figure 1 for Non-autoregressive Mandarin-English Code-switching Speech Recognition with Pinyin Mask-CTC and Word Embedding Regularization
Figure 2 for Non-autoregressive Mandarin-English Code-switching Speech Recognition with Pinyin Mask-CTC and Word Embedding Regularization
Figure 3 for Non-autoregressive Mandarin-English Code-switching Speech Recognition with Pinyin Mask-CTC and Word Embedding Regularization
Figure 4 for Non-autoregressive Mandarin-English Code-switching Speech Recognition with Pinyin Mask-CTC and Word Embedding Regularization
* 5 pages, 1 figure, submitted to INTERSPEECH2021 
  Access Paper or Ask Questions

Towards Lifelong Learning of End-to-end ASR


Apr 04, 2021
Heng-Jui Chang, Hung-yi Lee, Lin-shan Lee

Figure 1 for Towards Lifelong Learning of End-to-end ASR
Figure 2 for Towards Lifelong Learning of End-to-end ASR
Figure 3 for Towards Lifelong Learning of End-to-end ASR
Figure 4 for Towards Lifelong Learning of End-to-end ASR
* Submitted to INTERSPEECH 2021 
  Access Paper or Ask Questions

Auto-KWS 2021 Challenge: Task, Datasets, and Baselines


Mar 31, 2021
Jingsong Wang, Yuxuan He, Chunyu Zhao, Qijie Shao, Wei-Wei Tu, Tom Ko, Hung-yi Lee, Lei Xie

Figure 1 for Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
Figure 2 for Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
Figure 3 for Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
Figure 4 for Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
* 5 pages, 2 figures 
  Access Paper or Ask Questions

Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech


Mar 20, 2021
Chung-Ming Chien, Jheng-Hao Lin, Chien-yu Huang, Po-chun Hsu, Hung-yi Lee

Figure 1 for Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech
Figure 2 for Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech
Figure 3 for Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech
Figure 4 for Investigating on Incorporating Pretrained and Learnable Speaker Representations for Multi-Speaker Multi-Style Text-to-Speech
* Accepted by ICASSP 2021, in the special session of ICASSP 2021 M2VoC Challenge 
  Access Paper or Ask Questions

Adversarial defense for automatic speaker verification by cascaded self-supervised learning models


Feb 14, 2021
Haibin Wu, Xu Li, Andy T. Liu, Zhiyong Wu, Helen Meng, Hung-yi Lee

Figure 1 for Adversarial defense for automatic speaker verification by cascaded self-supervised learning models
Figure 2 for Adversarial defense for automatic speaker verification by cascaded self-supervised learning models
Figure 3 for Adversarial defense for automatic speaker verification by cascaded self-supervised learning models
Figure 4 for Adversarial defense for automatic speaker verification by cascaded self-supervised learning models
* Accepted to ICASSP 2021 
  Access Paper or Ask Questions

Pre-Training a Language Model Without Human Language


Dec 22, 2020
Cheng-Han Chiang, Hung-yi Lee

Figure 1 for Pre-Training a Language Model Without Human Language
Figure 2 for Pre-Training a Language Model Without Human Language
Figure 3 for Pre-Training a Language Model Without Human Language
Figure 4 for Pre-Training a Language Model Without Human Language
* 9 pages, work in progress 
  Access Paper or Ask Questions

Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis


Nov 17, 2020
Chung-Ming Chien, Hung-yi Lee

Figure 1 for Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis
Figure 2 for Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis
Figure 3 for Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis
Figure 4 for Hierarchical Prosody Modeling for Non-Autoregressive Speech Synthesis
* Accepted by SLT 2021 
  Access Paper or Ask Questions

Self-supervised Pre-training Reduces Label Permutation Instability of Speech Separation


Oct 29, 2020
Sung-Feng Huang, Shun-Po Chuang, Da-Rong Liu, Yi-Chen Chen, Gene-Ping Yang, Hung-yi Lee

Figure 1 for Self-supervised Pre-training Reduces Label Permutation Instability of Speech Separation
Figure 2 for Self-supervised Pre-training Reduces Label Permutation Instability of Speech Separation
Figure 3 for Self-supervised Pre-training Reduces Label Permutation Instability of Speech Separation
Figure 4 for Self-supervised Pre-training Reduces Label Permutation Instability of Speech Separation
* submitted to ICASSP2021 
  Access Paper or Ask Questions

Pretrained Language Model Embryology: The Birth of ALBERT


Oct 29, 2020
Cheng-Han Chiang, Sung-Feng Huang, Hung-yi Lee

Figure 1 for Pretrained Language Model Embryology: The Birth of ALBERT
Figure 2 for Pretrained Language Model Embryology: The Birth of ALBERT
Figure 3 for Pretrained Language Model Embryology: The Birth of ALBERT
Figure 4 for Pretrained Language Model Embryology: The Birth of ALBERT
* Accepted to EMNLP 2020, short paper 
  Access Paper or Ask Questions

FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention


Oct 27, 2020
Yist Y. Lin, Chung-Ming Chien, Jheng-Hao Lin, Hung-yi Lee, Lin-shan Lee

Figure 1 for FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention
Figure 2 for FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention
Figure 3 for FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention
Figure 4 for FragmentVC: Any-to-Any Voice Conversion by End-to-End Extracting and Fusing Fine-Grained Voice Fragments With Attention
* Submitted to ICASSP 2021, equal contribution from first two authors 
  Access Paper or Ask Questions

Language Representation in Multilingual BERT and its applications to improve Cross-lingual Generalization


Oct 23, 2020
Chi-Liang Liu, Tsung-Yuan Hsu, Yung-Sung Chuang, Chung-Yi Li, Hung-yi Lee

Figure 1 for Language Representation in Multilingual BERT and its applications to improve Cross-lingual Generalization
Figure 2 for Language Representation in Multilingual BERT and its applications to improve Cross-lingual Generalization
Figure 3 for Language Representation in Multilingual BERT and its applications to improve Cross-lingual Generalization
Figure 4 for Language Representation in Multilingual BERT and its applications to improve Cross-lingual Generalization
* preprint 
  Access Paper or Ask Questions

Unsupervised Deep Learning based Multiple Choices Question Answering: Start Learning from Basic Knowledge


Oct 21, 2020
Chi-Liang Liu, Hung-yi Lee

Figure 1 for Unsupervised Deep Learning based Multiple Choices Question Answering: Start Learning from Basic Knowledge
Figure 2 for Unsupervised Deep Learning based Multiple Choices Question Answering: Start Learning from Basic Knowledge
Figure 3 for Unsupervised Deep Learning based Multiple Choices Question Answering: Start Learning from Basic Knowledge
Figure 4 for Unsupervised Deep Learning based Multiple Choices Question Answering: Start Learning from Basic Knowledge
* preprint 
  Access Paper or Ask Questions

What makes multilingual BERT multilingual?


Oct 20, 2020
Chi-Liang Liu, Tsung-Yuan Hsu, Yung-Sung Chuang, Hung-yi Lee

Figure 1 for What makes multilingual BERT multilingual?
Figure 2 for What makes multilingual BERT multilingual?
Figure 3 for What makes multilingual BERT multilingual?
Figure 4 for What makes multilingual BERT multilingual?
* arXiv admin note: substantial text overlap with arXiv:2004.09205 
  Access Paper or Ask Questions

Language Representation in Multilingual BERTand its applications to improve Cross-lingual Generalization


Oct 20, 2020
Chi-Liang Liu, Tsung-Yuan Hsu, Yung-Sung Chuang, Hung-yi Lee

Figure 1 for Language Representation in Multilingual BERTand its applications to improve Cross-lingual Generalization
Figure 2 for Language Representation in Multilingual BERTand its applications to improve Cross-lingual Generalization
Figure 3 for Language Representation in Multilingual BERTand its applications to improve Cross-lingual Generalization
Figure 4 for Language Representation in Multilingual BERTand its applications to improve Cross-lingual Generalization

  Access Paper or Ask Questions

TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech


Jul 12, 2020
Andy T. Liu, Shang-Wen Li, Hung-yi Lee

Figure 1 for TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
Figure 2 for TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
Figure 3 for TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
Figure 4 for TERA: Self-Supervised Learning of Transformer Encoder Representation for Speech
* Submitted to IEEE TASLP, currently under review 
  Access Paper or Ask Questions