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 Siyuan Feng

Siyuan Feng

Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation


Apr 02, 2021
Siyuan Feng, Piotr Żelasko, Laureano Moro-Velázquez, Odette Scharenborg

Figure 1 for Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation
Figure 2 for Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation
Figure 3 for Unsupervised Acoustic Unit Discovery by Leveraging a Language-Independent Subword Discriminative Feature Representation
* Submitted to INTERSPEECH 2021 
  Access Paper or Ask Questions

Quantifying Bias in Automatic Speech Recognition


Apr 01, 2021
Siyuan Feng, Olya Kudina, Bence Mark Halpern, Odette Scharenborg

Figure 1 for Quantifying Bias in Automatic Speech Recognition
Figure 2 for Quantifying Bias in Automatic Speech Recognition
Figure 3 for Quantifying Bias in Automatic Speech Recognition
Figure 4 for Quantifying Bias in Automatic Speech Recognition
* Submitted to INTERSPEECH (IS) 2021. This preprint version differs slightly from the version submitted to IS 2021: Figure 1 is not included in IS 2021 
  Access Paper or Ask Questions

The effectiveness of unsupervised subword modeling with autoregressive and cross-lingual phone-aware networks


Dec 17, 2020
Siyuan Feng, Odette Scharenborg

Figure 1 for The effectiveness of unsupervised subword modeling with autoregressive and cross-lingual phone-aware networks
Figure 2 for The effectiveness of unsupervised subword modeling with autoregressive and cross-lingual phone-aware networks
Figure 3 for The effectiveness of unsupervised subword modeling with autoregressive and cross-lingual phone-aware networks
Figure 4 for The effectiveness of unsupervised subword modeling with autoregressive and cross-lingual phone-aware networks
* 18 pages (including 1 page for supplementary material), 13 figures. Submitted to IEEE Open Journal of Signal Processing (OJ-SP) 
  Access Paper or Ask Questions

Joint predictions of multi-modal ride-hailing demands: a deep multi-task multigraph learning-based approach


Nov 11, 2020
Jintao Ke, Siyuan Feng, Zheng Zhu, Hai Yang, Jieping Ye

Figure 1 for Joint predictions of multi-modal ride-hailing demands: a deep multi-task multigraph learning-based approach
Figure 2 for Joint predictions of multi-modal ride-hailing demands: a deep multi-task multigraph learning-based approach
Figure 3 for Joint predictions of multi-modal ride-hailing demands: a deep multi-task multigraph learning-based approach
Figure 4 for Joint predictions of multi-modal ride-hailing demands: a deep multi-task multigraph learning-based approach

  Access Paper or Ask Questions

Unsupervised Pattern Discovery from Thematic Speech Archives Based on Multilingual Bottleneck Features


Nov 03, 2020
Man-Ling Sung, Siyuan Feng, Tan Lee

Figure 1 for Unsupervised Pattern Discovery from Thematic Speech Archives Based on Multilingual Bottleneck Features
Figure 2 for Unsupervised Pattern Discovery from Thematic Speech Archives Based on Multilingual Bottleneck Features
Figure 3 for Unsupervised Pattern Discovery from Thematic Speech Archives Based on Multilingual Bottleneck Features
Figure 4 for Unsupervised Pattern Discovery from Thematic Speech Archives Based on Multilingual Bottleneck Features
* 8 pages, accepted and presented in APSIPA-APC 2018. This work was done when Man-Ling Sung and Siyuan Feng were postgraduate students in the Chinese University of Hong Kong 
  Access Paper or Ask Questions

Show and Speak: Directly Synthesize Spoken Description of Images


Oct 23, 2020
Xinsheng Wang, Siyuan Feng, Jihua Zhu, Mark Hasegawa-Johnson, Odette Scharenborg

Figure 1 for Show and Speak: Directly Synthesize Spoken Description of Images
Figure 2 for Show and Speak: Directly Synthesize Spoken Description of Images
Figure 3 for Show and Speak: Directly Synthesize Spoken Description of Images
Figure 4 for Show and Speak: Directly Synthesize Spoken Description of Images

  Access Paper or Ask Questions

How Phonotactics Affect Multilingual and Zero-shot ASR Performance


Oct 22, 2020
Siyuan Feng, Piotr Żelasko, Laureano Moro-Velázquez, Ali Abavisani, Mark Hasegawa-Johnson, Odette Scharenborg, Najim Dehak

Figure 1 for How Phonotactics Affect Multilingual and Zero-shot ASR Performance
Figure 2 for How Phonotactics Affect Multilingual and Zero-shot ASR Performance
Figure 3 for How Phonotactics Affect Multilingual and Zero-shot ASR Performance
Figure 4 for How Phonotactics Affect Multilingual and Zero-shot ASR Performance
* Submitted to ICASSP 2021. The first 2 authors contributed equally to this work 
  Access Paper or Ask Questions

Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling


Aug 06, 2020
Siyuan Feng, Odette Scharenborg

Figure 1 for Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling
Figure 2 for Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling
Figure 3 for Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling
Figure 4 for Unsupervised Subword Modeling Using Autoregressive Pretraining and Cross-Lingual Phone-Aware Modeling
* 5 pages, 3 figures. Accepted for publication in INTERSPEECH 2020, Shanghai, China 
  Access Paper or Ask Questions

Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages


Jul 29, 2020
Siyuan Feng

Figure 1 for Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages
Figure 2 for Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages
Figure 3 for Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages
Figure 4 for Exploiting Cross-Lingual Knowledge in Unsupervised Acoustic Modeling for Low-Resource Languages
* Ph.D. Thesis Submitted in May 2020 in partial fulfilment of the requirements for the Degree of Doctor of Philosophy in Electronic Engineering, The Chinese University of Hong Kong (CUHK) 134 pages 
  Access Paper or Ask Questions

Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal


Oct 30, 2019
Zhiyuan Peng, Siyuan Feng, Tan Lee

Figure 1 for Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal
Figure 2 for Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal
Figure 3 for Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal
Figure 4 for Mixture factorized auto-encoder for unsupervised hierarchical deep factorization of speech signal

  Access Paper or Ask Questions

Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling


Aug 09, 2019
Siyuan Feng, Tan Lee

Figure 1 for Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling
Figure 2 for Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling
Figure 3 for Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling
Figure 4 for Exploiting Cross-Lingual Speaker and Phonetic Diversity for Unsupervised Subword Modeling
* 12 pages, 6 figures. This manuscript has been accepted for publication as a regular paper in the IEEE Transactions on Audio, Speech and Language Processing 
  Access Paper or Ask Questions

Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling


Aug 09, 2019
Siyuan Feng, Tan Lee, Zhiyuan Peng

Figure 1 for Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling
Figure 2 for Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling
Figure 3 for Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling
Figure 4 for Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling
* 5 pages, 3 figures, accepted for publication in INTERSPEECH 2019, Graz, Austria 
  Access Paper or Ask Questions

Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation


Jul 03, 2019
Siyuan Feng, Tan Lee

Figure 1 for Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation
Figure 2 for Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation
Figure 3 for Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation
Figure 4 for Improving Unsupervised Subword Modeling via Disentangled Speech Representation Learning and Transformation
* 5 pages, 3 figures, accepted for publication in INTERSPEECH 2019, Graz, Austria 
  Access Paper or Ask Questions

CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario


May 13, 2019
Huichu Zhang, Siyuan Feng, Chang Liu, Yaoyao Ding, Yichen Zhu, Zihan Zhou, Weinan Zhang, Yong Yu, Haiming Jin, Zhenhui Li

Figure 1 for CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
Figure 2 for CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
Figure 3 for CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
* WWW 2019 Demo Paper 
  Access Paper or Ask Questions