SHENG LI (李 勝) English version

テニュアトラック 研究員

国立研究開発法人情報通信研究機構 (NICT)
└──ユニバーサルコミュニケーション研究所 (UCRI)
  └──先進的音声翻訳研究開発推進センター (ASTREC)
    └──先進的音声技術研究室(ASTL)

現在の研究

• 次世代音声翻訳・多言語音声認識・合成処理技術の研究開発.

• Welcome research students/visiting researchers for long and short-time visits.
  I am expecting to collaborate with intelligent brains all over the world.
  

学歴

QS2021 world ranking (東アジアの大学ランキング ) (清華/北京大/交/浙/復/科/南京大の卒業生は北京に直接定住することができます)

• 2002年7月 中国江蘇省無錫(むしゃく)市第一高等学校(高校部) 卒業

• 2006年7月 南京大学（国立中央大学，1949年国立南京大学と改称）学士修了(理学)

• 2009年7月 南京大学大学院修士課程修了 (中国科学院，香港中文大学，南京大学連携項目課程同期修了)

• 2016年3月 京都大学大学院情報学研究科知能情報学専攻博士後期課程修了 (京都大学国費留学生特別配置10月入学, 京都大学推薦, 非CSC)
  取得学位: 博士(情報学); 学位授与番号: 甲第19849号

職歴

• 2009.8 中国科学院 深セン市 研究員（2012.3まで）
  メディア処理技術を用いた外国語学習（NSFC60772165）, 聴覚障害者支援研究（NSFC61135003）に従事

• 2012.4 Sogou [株，北京市] 研究員（2012.9まで）
  音声認識ピン音入力方法 (音響モデルの音素誤り最小化学習)

• 2012.10 京都大学音声メディア研究室（2017.1まで）
  1. DNN-HMM音声認識の活用による講演・講義の字幕付与に従事 (MEXT PhD 2016.3まで）
  2. 音声認識へのDNN-HMMの適用と話者適応（NICT 協力研究 2016.3まで）
  3. ERATOプロジェクトに従事（ERATO研究員 2016.4 -- 2017.1まで）

• 2017.1 情報通信研究機構NICT,先進的音声技術研究室ASTL (現在に至る)
  1. Tokyo Olympic 2020多言語音声認識・合成処理技術の研究開発（研究員 2020.3まで）
  2. 次世代音声認識技術の研究開発に従事 (テニュアトラック研究員 2020.4 現在に至る)

• 2019.4 オックスフォード大学、コンピュータサイエンス、客員研究員（2019.5まで）
  Trust AI and explainable deep neural network

• 2021.12 京都大学大学院 情報学研究科 修士課程アドバイザー (現在に至る)
  セキュリティ, 音声処理

• 2022.4 シンガポール 南洋理工大学 客員研究員 (現在に至る)
  音声認識

受賞

• NICT
  2022年 Co-1st place in Main/OOD tracks in INTERSPEECH2022 special session: VoiceMOS challenge
  2021年 3rd/4th place in constrained/unconstrained resource multilingual ASR tracks of OLR2021 challenge
  2021年 Supervised student (Soky Kak) got best student paper nomination in O-COCOSDA2021
  2021年 令和３年度情報通信研究機構表彰: 成績優秀表彰優秀賞（団体)
  2020年 Two supervised students (Hao Shi and Yuqin Lin) got ISCA-grants in INTERSPEECH2020
  2020年 Supervised student (Yaowei Han) got best student paper nomination in IEEE-ICME2020

• Kyoto University
  2018年 第34回 電気通信普及財団 テレコムシステム技術学生賞
  2016年 Paper nominated as ACM/IEEE Trans. Audio, Speech and Language Process. cover
  2012年-2016年 京都大学入学料・授業料の全部免除
  2012年 日本政府（文部科学省）奨学金, 京都大学推薦国費留学生特別配置入学

• Joint Lab CAS and CUHK
  2012年 ポートランド, USA，INTERSPEECH会議へIBM 旅行補助賞金
  2011年 香港青年起業家プログラムの創造的な企画賞
  2011年 中国科学院職員優秀賞

• Nanjing University
  2004年 南京大学勵志奨学金
  2002年 香港陳蔭川財団大学新入生優秀者奨学金
  2002年 中国江蘇省化学オリンピック二等賞, 生物学オリンピック三等賞

競争的資金等

• NICT international funding (PI): 2022-2024 (ongoing)
  Bridging Eurasia from Sea -- Multilingual Speech Recognition for Maritime Silkroad

• Grant-in-Aid for Young Scientists (PI): 2021-2023 (ongoing)
  Phantom in the Opera -- the Vulnerabilities of Speech Interface for Robotic Dialogue System

• NICT tenure-track start-up funding (PI): 2020-2023 (ongoing)
  Advanced Multilingual End-to-End Speech Recognition

• NICT international funding (PI): 2020-2022
  Bridging Eurasia -- Multilingual Speech Recognition for Silkroad

• NII Open Collaborative Research (collaborator): 2020-2021
  Speaker De-identification with Provable Privacy in Speech Data Release

• Grant-in-Aid for Research Activity Start-up (PI): 2019-2021
  Next generation multilingual End-to-End speech recognition (from G30 to G200)

発表文献

(13 ICASSP, 22 INTERSPEECH, 3 Odyssey, 1 ASRU, 1 SIGDIAL, 2 SLT, 1 ICME, 2 TASLP, 1 SPEECH COMMUN, 1 EURASIP JASMP)

博士論文

• Sheng Li (李 勝).
  Speech Recognition Enhanced by Lightly-supervised and Semi-supervised Acoustic Model Training.
  Ph.D. Thesis, Kyoto University, 2016.
  学位論文: 音響モデルの準教師付き及び半教師付き学習による音声認識，指導教授: 河原達也先生，Feb. 2016.

著書

• S. Li, Voices of the Himalayas: Investigation of Speech Recognition Technology for the Tibetan Language, ISBN**, 2022.

• S. Li, Phantom in the Opera: The Vulnerabilities of Speech-based Artificial Intelligence Systems, ISBN978-4-904020, 2022.

• X. Lu, S. Li, M. Fujimoto, Speech-to-Speech Translation, pp. 21-38, Springer Singapore, 2020.
  S. Li, Chapter: From Shallow to Deep and Very Deep.
  S. Li, Chapter: End-to-End and CTC models.

招待講演

• Phoneme-level articulatory animation in pronunciation training using EMA data,
  2012, Speech Synthesis Lab., Tsinghua University, host: Prof. Zhiyong Wu.

• Lightly-supervised training and confidence estimation by using CRF classifiers,
  2014, Speech and Cognition Lab., Tianjin University, host: Prof. Jianwu Dang and Prof. Kiyoshi Honda.

• End-to-End Speech Recognition,
  2019, University of Tokyo, host: Dr. Jinze Yu.

学術論文誌 掲載論文 (Peer reviewed)

• Kak Soky, Masato Mimura, Chenhui Chu, Tatsuya Kawahara, S. Li, Chenchen Ding, Sethserey Sam.
  TriECCC: Trilingual Corpus of the Extraordinary Chambers in the Courts of Cambodia for Speech Recognition and Translation Studies.
  International Journal of Asian Language Processing, 2022. (invited paper for best student paper nominated in OCOCOSDA-2021)

• C. Fan, H. Zhang, J. Yi, Z. Lv, J. Tao, T. Li, G. Pei, X. Wu, S. Li.
  SpecMNet: Spectrum Mend Network for Monaural Speech Enhancement.
  Applied Acoustics, Volume 194, 15 June 2022, 108792.

• S. Shimizu, C. Chu, S. Li, S. Kurohashi.
  Cross-Lingual Transfer Learning for End-to-End Speech Translation.
  Journal of Natural Language Processing (JNLP), Vol.29, No.2, 2022 June.

• X. Chen, H. Huang, and S. Li,
  Adversarial Attack and Defense on Deep Neural Network-based Voice Processing Systems: An Overview.
  Applied Sciences, Special Issues of Machine Speech Communication, 2021. (Peer reviewed, invited survey paper)

• S. Qin, L. Wang, S. Li (corresponding), J. Dang and L. Pan.
  Improving Low-resource Tibetan End-to-end ASR by Multilingual and Multi-level Unit Modeling.
  EURASIP Journal on Audio, Speech and Music Processing. (EURASIP JASMP), No.2 , 2022.

• P. Shen, X. Lu, S. Li(patent co-inventor), H. Kawai.
  Knowledge Distillation-based Representation Learning for Short-Utterance Spoken Language Identification.
  IEEE Trans. Audio, Speech \& Language Process. (TASLP), vol. 28, pp. 2674--2683, 2020.

• S. Li, Y.Akita, and T.Kawahara.
  Semi-supervised acoustic model training by discriminative data selection from multiple ASR systems' hypotheses.
  IEEE Trans. Audio, Speech \& Language Process. (TASLP), Vol.24, No.9, pp.1520--1530, 2016.
  (Cover Paper, IEEE Signal Processing Society Japan Student Journal Paper Award)

• S. Li, Y.Akita, and T.Kawahara.
  Automatic lecture transcription based on discriminative data selection for lightly supervised acoustic model training.
  IEICE Trans., Vol.E98-D, No.8, pp.1545--1552, 2015.

• L. Wang, H. Chen, S. Li, and H. Meng.
  Phoneme-level articulatory animation in pronunciation training,
  Speech Communication (SPEECH COMMUN), Vol. 54, Issue 7, Sept. pp. 845--856, 2012.

国際会議 発表論文 (Peer reviewed)

2022 (Tenure-track @NICT)
• Zhuo Gong, Saito Daisuke, S. Li, Hisashi Kawai, Minematsu Nobuaki,
  Can We Train a Language Model Inside an End-to-End ASR Model? - Investigating Effective Implicit Language Modeling.
  in Proc. COLING2022 Workshop, 2022. (accept)

• Kak Soky, Zhuo Gong, S. Li,
  NICT-Tib1: A Public Speech Corpus of Lhasa Dialect for Benchmarking Tibetan Language Speech Recognition Systems.
  in Proc. O-COCOSDA, 2022. (accept)

• S. Li, Jiyi Li, Qianying Liu, Zhuo Gong,
  An End-to-End Chinese and Japanese Bilingual Speech Recognition Systems with Shared Character Decomposition.
  in Proc. ICONIP, 2022. (accept)

• Xiaojiao Chen, Hao Huang, S. Li,
  GVec: Extracting Speaker Embedding from End-to-End Speech Recognition Model.
  in Proc. ICONIP, 2022. (accept)

• Guangxing Li, Wangjin Zhou, S. Li, Yi Zhao, Jichen Yang, Hao Huang,
  Investigating Effective Domain Adaptation Method for Speaker Verification Task.
  in Proc. ICONIP, 2022. (accept)

• Hao Shi, Longbiao Wang, S. Li, Jianwu Dang, Tatsuya Kawahara,
  Subband-Based Spectrogram Fusion for Speech Enhancement by Combining Mapping and Masking Approaches.
  in Proc. APSIPA ASC, 2022. (accept)

• Longfei Yang, Jiyi Li, S. Li and Takahiro Shinozaki,
  Multi-Domain Dialogue State Tracking with Top-k Slot Self Attention.
  in Proc. SIGdial Meeting Discourse \& Dialogue, 2022. (accept)

• Kak Soky, S. Li, Masato Mimura, Chenhui Chu and Tatsuya Kawahara,
  Leveraging Simultaneous Translation for Enhancing Transcription of Low-resource Language via Cross Attention Mechanism.
  in Proc. INTERSPEECH, 2022. (accept)

• Longfei Yang, Wenqing Wei, S. Li, Jiyi Li and Takahiro Shinozaki,
  Augmented Adversarial Self-Supervised Learning for Early-Stage Alzheimer's Speech Detection.
  in Proc. INTERSPEECH, 2022. (accept)

• Kai Li, S. Li, Xugang Lu, Masato Akagi, Meng Liu, Lin Zhang, Chang Zeng, Longbiao Wang, Jianwu Dang and Masashi Unoki,
  Data Augmentation Using McAdams-Coefficient-Based Speaker Anonymization for Fake Audio Detection.
  in Proc. INTERSPEECH, 2022. (accept)

• Zhengdong Yang, Wangjin Zhou, Chenhui Chu, S. Li, Raj Dabre, Raphael Rubino and Yi Zhao,
  Fusion of Self-supervised Learned Models for MOS Prediction.
  in Proc. INTERSPEECH, 2022. (accept)

• Siqing Qin, Longbiao Wang, S. Li, Yuqin Lin and Jianwu Dang,
  Finer-grained Modeling units-based Meta-Learning for Low-resource Tibetan Speech Recognition.
  in Proc. INTERSPEECH, 2022. (accept)

• Hao Shi, Longbiao Wang, S. Li, Jianwu Dang and Tatsuya Kawahara,
  Monaural Speech Enhancement Based on Spectrogram Decomposition for Convolutional Neural Network-sensitive Feature Extraction.
  in Proc. INTERSPEECH, 2022. (accept)

• Nan Li, Meng Ge, Longbiao Wang, Masashi Unoki, S. Li and Jianwu Dang,
  Global Signal-to-noise Ratio Estimation Based on Multi-subband Processing Using Convolutional Neural Network.
  in Proc. INTERSPEECH, 2022. (accept)

• Kai Li, Xugang Lu, Masato Akagi, Jianwu Dang, S. Li, Masashi Unoki,
  Relationship Between Speakers' Physiological Structure and Acoustic Speech Signals:Data-Driven Study Based on Frequency-Wise Attentional Neural Network
  In Proc. EUSIPCO (European Signal Processing Conference), 2022. (accepted)

• Zhuo Gong, Daisuke Saito, Longfei Yang, Takahiro Shinozaki, S. Li, Hisashi Kawai and Nobuaki Minematsu,
  Self-Adaptive Multilingual ASR Rescoring with Language Identification and Unified Language Model.
  In Proc. ISCA-Odyssey (The Speaker and Language Recognition Workshop), 2022. (accepted)

• S. Li, J. Li, Q. Liu, Z. Gong,
  Adversarial Speech Generation and Natural Speech Recovery for Speech Content Protection.
  in Proc. LREC (Language Resources and Evaluation Conference), 2022. (accepted)

• Y. Lv, L. Wang, M. Ge, S. Li (corresponding), C. Ding, L. Pan, Y. Wang, J. Dang, K. Honda,
  Compressing Transformer-based ASR Model by Task-driven Loss and Attention-based Multi-level Feature Distillation.
  in Proc. IEEE-ICASSP, pp. 7992--7996, 2022.

• K. Wang, Y. Peng, H. Huang, Y. Hu, and S. Li,
  Mining Hard Samples Locally and Globally for Improved Speech Separation.
  in Proc. IEEE-ICASSP, pp. 6037--6041, 2022.

2021 (Tenure-track @NICT)
• D. Liu, L. Wang, S. Li, H. Li, C. Ding, J. Zhang and J. Dang.
  Exploring Effective Speech Representation via ASR for High-Quality End-to-End Multispeaker TTS.
  In Proc. ICONIP, 2021. (accepted)

• L. Qiang, H. Shi, M. Ge, H. Yin, N. Li, L. Wang, S. Li and J. Dang.
  Speech Dereverberation Based on Scale-aware Mean Square Error Loss.
  In Proc. ICONIP, 2021. (accepted)

• H. Yin, L. Qiang, H. Shi, L. Wang, S. Li, M. Ge, G. Zhang and J. Dang.
  Simultaneous Progressive Filtering-based Monaural Speech Enhancement.
  In Proc. ICONIP, 2021. (accepted)

• K. Soky, M. Mimura, T. Kawahara, S. Li, C. Ding, C. Chu, and S. Sam.
  Khmer Speech Translation Corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC).
  In Proc. O-COCOSDA, pp. 122--127, 2021. (Best student paper nomination, invited as fast tracking journal paper)

• H. Shi, L. Wang, S. Li, C. Fan, J. Dang, and T. Kawahara.
  Spectrograms Fusion-based End-to-End Robust Automatic Speech Recognition.
  In Proc. APSIPA ASC, pp. 438--442, 2021.

• Y. Peng, J. Zhang, H. Zhang, H. Xu, H. Huang, S. Li, and E.S. Chng.
  Multilingual Approach to Joint Speech and Accent Recognition with DNN-HMM Framework.
  In Proc. APSIPA ASC, pp. 1043-1048, 2021.

• K. Soky, S. Li, M. Mimura, C. Chu, and T. Kawahara.
  On the Use of Speaker Information for Automatic Speech Recognition in Speaker-imbalanced Corpora.
  In Proc. APSIPA ASC, pp. 433-437, 2021.

• D. Wang, S. Ye, X. Hu, S. Li, and X. Xu,
  An End-to-End Dialect Identification System with Transfer Learning from a Multilingual Automatic Speech Recognition Model.
  in Proc. INTERSPEECH, pp. 3266--3270, 2021.

• K. Wang, H. Huang, Y. Hu, Z. Huang, and S. Li,
  End-to-End Speech Separation Using Orthogonal Representation in Complex and Real Time- Frequency Domain.
  in Proc. INTERSPEECH, pp. 3046--3050, 2021.

• N. Li, L. Wang, M. Unoki, S. Li, R. Wang, M. Ge, and J. Dang,
  Robust voice activity detection using a masked auditory encoder based convolutional neural network.
  in Proc. IEEE-ICASSP, pp. 6828--6832, 2021.

• S. Chen, X. Hu, S. Li, and X. Xu,
  An investigation of using hybrid modeling units for improving End-to-End speech recognition systems.
  in Proc. IEEE-ICASSP, pp. 6743--6747, 2021.

• H. Huang, K. Wang, Y. Hu, and S. Li,
  Encoder-Decoder based pitch tracking and joint model training for Mandarin tone classification.
  in Proc. IEEE-ICASSP, pp. 6943--6947, 2021.

2020 (Tenure-track @NICT)
• A. Thida, N. Han, S. Oo, S. Li, and C. Ding.
  VOIS: The First Speech Therapy App in the World for Myanmar Hearing-Impaired Children.
  In Proc. O-COCOSDA, pp. 151--154, 2020.

• S. Guo, L. Wang, S. Li, J. Zhang, C. Gong, Y. Wang, J. Dang, and K. Honda.
  Investigation of Effectively Synthesizing Code-switched Speech Using Highly Imbalanced Mix-lingual Data.
  In Proc. ICONIP, pp. 36--47, 2020.

• Y. Lin, L. Wang, S. Li, J. Dang, and C. Ding.
  Staged Knowledge Distillation for End-to-End Dysarthric Speech Recognition and Speech Attribute Transcription.
  In Proc. INTERSPEECH, pp. 4791--4795, 2020 (Travel Granted by ISCA).

• H. Shi, L. Wang, S. Li, C. Ding, M. Ge, N. Li, J. Dang, and H. Seki.
  Singing Voice Extraction with Attention based Spectrograms Fusion.
  In Proc. INTERSPEECH, pp. 2412--2416, 2020 (Travel Granted by ISCA).

• S. Li, X. Lu, R. Dabre, P. Shen, and H. Kawai.
  Joint Training End-to-End Speech Recognition Systems with Speaker Attributes.
  In Proc. ISCA-Odyssey (The Speaker and Language Recognition Workshop), pp. 385--390, 2020.

• P. Shen, X. Lu, K. Sugiura, S. Li, and H. Kawai.
  Compensation on x-vector for short utterance spoken language identification.
  In Proc. ISCA-Odyssey (The Speaker and Language Recognition Workshop), pp. 47--52, 2020.

• Y. Han, S. Li, Y. Cao, Q. Ma, and M. Yoshikawa.
  Voice-Indistinguishability: Protecting Voiceprint in Privacy Preserving Speech Data Release.
  In Proc. IEEE-ICME, pp. 1--6, 2020. (Best student paper nomination, select as fast tracking journal paper of IEEE Trans. Multimedia (TMM))

• Y. Lin, L. Wang, J. Dang, S. Li, and C. Ding.
  End-To-End Articulatory Modeling for Dysarthria Articulatory Attribute Detection.
  In Proc. IEEE-ICASSP, pp. 7349--7353, 2020.

• H. Shi, L. Wang, M. Ge, S. Li, and J. Dang.
  Spectrograms Fusion with Minimum Difference Masks Estimation for Monaural Speech Dereverberation.
  In Proc. IEEE-ICASSP, pp. 7544--7548, 2020.

2019 (Post-doc @NICT)
• K.Soky, S. Li, T.Kawahara, and S.Seng.
  Multi-lingual transformer training for Khmer automatic speech recognition.
  In Proc. APSIPA ASC, pp. 1893--1896, 2019.

• L. Pan, S. Li, L. Wang, and J. Dang.
  Effective training End-to-End ASR systems for low-resource Lhasa dialect of Tibetan language.
  In Proc. APSIPA ASC, pp. 1152--1156, 2019.

• X. Lu, P. Shen, S. Li, Y. Tsao, and H. Kawai.
  Class-wise Centroid Distance Metric Learning for Acoustic Event Detection.
  In Proc. INTERSPEECH, pp. 3614--3618, 2019.

• S. Li, X. Lu, C. Ding, P. Shen, T. Kawahara, and H. Kawai.
  Investigating Radical-based End-to-End Speech Recognition Systems for Chinese Dialects and Japanese.
  In Proc. INTERSPEECH, pp. 2200--2204, 2019.

• S. Li, C. Ding, X. Lu, P. Shen, T. Kawahara, and H. Kawai.
  End-to-End Articulatory Attribute Modeling for Low-resource Multilingual Speech Recognition.
  In Proc. INTERSPEECH, pp. 2145--2149, 2019.

• S. Li, R. Dabre, X. Lu, P. Shen, T. Kawahara, and H. Kawai.
  Improving Transformer-based Speech Recognition Systems with Compressed Structure and Speech Attributes Augmentation.
  In Proc. INTERSPEECH, pp. 4400--4404, 2019.

• P.Shen, X.Lu, S. Li, and H.Kawai.
  Interactive learning of teacher-student model for short utterance spoken language identification.
  In Proc. IEEE-ICASSP, pp. 5981--5985, 2019.

• R.Takashima, S. Li, and H.Kawai.
  Investigation of Sequence-level Knowledge Distillation Methods for CTC Acoustic Models.
  In Proc. IEEE-ICASSP, pp. 6156--6160, 2019.

2018 (Post-doc @NICT)
• S. Li, X.Lu, R.Takashima, P.Shen, T.Kawahara, and H.Kawai.
  Improving very deep time-delay neural network with vertical-attention for effectively training CTC-based ASR systems.
  In Proc. IEEE Spoken Language Technology Workshop (IEEE-SLT), pp. 77--83, 2018.

• S. Li, X.Lu, R.Takashima, P.Shen, T.Kawahara, and H.Kawai.
  Improving CTC-based Acoustic Model with Very Deep Residual Time-delay Neural Networks.
  In Proc. INTERSPEECH, pp. 3708--3712, 2018.

• P.Shen, X.Lu, S. Li, and H.Kawai.
  Feature Representation of Short Utterances based on Knowledge Distillation for Spoken Language Identification.
  In Proc. INTERSPEECH, pp. 1813--1817, 2018.

• X.Lu, P.Shen, S. Li, Y.Tsao, and H.Kawai.
  Temporal Attentive Pooling for Acoustic Event Detection.
  In Proc. INTERSPEECH, pp. 1354--1357, 2018.

• R.Takashima, S. Li, and H.Kawai.
  An Investigation of a Knowledge Distillation Method for CTC Acoustic Models.
  In Proc. IEEE-ICASSP, pp. 5809--5813, 2018.

• R.Takashima, S. Li, and H.Kawai.
  CTC Loss Function with a Unit-level Ambiguity Penalty.
  In Proc. IEEE-ICASSP, pp. 5909--5913, 2018.

2017 (Post-doc @NICT)
• S. Li, X.Lu, P.Shen, R.Takashima, T.Kawahara, and H.Kawai.
  Incremental training and constructing the very deep convolutional residual network acoustic models.
  In Proc. IEEE Workshop Automatic Speech Recognition \& Understanding (IEEE-ASRU), pp. 222--227, 2017.

• P. Shen, X. Lu, S. Li, and H. Kawai.
  Conditional Generative Adversarial Nets Classifier for Spoken Language Identification.
  In Proc. INTERSPEECH, pp. 2814--2818, 2017.

• S. Li, X.Lu, S.Sakai, M.Mimura, and T.Kawahara.
  Semi-supervised ensemble DNN acoustic model training.
  In Proc. IEEE-ICASSP, pp. 5270--5274, 2017.

before 2017 (Kyoto Univ.)
• S. Li, X.Lu, S.Mori, Y.Akita, and T.Kawahara.
  Confidence Estimation for Speech Recognition Systems using Conditional Random Fields Trained with Partially Annotated Data
  Int'l Sympo. Chinese Spoken Language Processing (ISCSLP), pp.1--5, 2016.

• S. Li, Y.Akita, and T.Kawahara.
  Data selection from multiple ASR systems' hypotheses for unsupervised acoustic model training.
  In Proc. IEEE-ICASSP, pp. 5875--5879, 2016.

• S. Li, Y.Akita, and T.Kawahara.
  Discriminative data selection for lightly supervised training of acoustic model using closed caption texts.
  In Proc. INTERSPEECH, pp. 3526--3530, 2015.(oral)

• S. Li, X.Lu, Y.Akita, and T.Kawahara.
  Ensemble speaker modeling using speaker adaptive training deep neural network for speaker adaptation.
  In Proc. INTERSPEECH, pp. 2892--2896, 2015.

• S. Li, Y.Akita, and T.Kawahara.
  Corpus and transcription system of Chinese Lecture Room.
  In Proc. Int'l Sympo. Chinese Spoken Language Processing (ISCSLP), pp. 442--445, 2014.

before 2013 (Joint Lab. CAS & CUHK)
• S. Li and L. Wang.
  Cross Linguistic Comparison of Mandarin and English EMA Articulatory Data,
  In Proc. INTERSPEECH, pp. 903--906, 2012. (Travel granted by IBM research)

• S. Li, L. Wang, and E. Qi.
  The Phoneme-level Articulator Dynamics for Pronunciation Animation,
  In Proc. IALP, pp. 283--286, 2011.

• J. Chen, L. Wang, C. Li, J. Hu, and S. Li.
  IELS: A Computer-aided Pronunciation Training System for Undergraduate Students,
  ICETC, Vol.1, pp. 338--342, 2010.

Challenges・Demo

• X. Chen, G. Li, H. Huang, W. Zhou, S. Li, Y. Cao, Y. Zhao.
  VoicePrivacy Challenge: System description.
  In VoicePrivacy 2022 Challenge Workshop (Interspeech2022)

• Guangxing Li, Wangjin Zhou, S. Li, Yi Zhao, Hao Huang, Jichen Yang.
  System Description for the CN-Celeb Speaker Recognition Challenge 2022.
  In CNSRC (the CN-Celeb Speaker Recognition Challenge), Speaker Odyssey 2022

• S. Li, R. Dabre, R. Raphael, W. Zhou, Z. Yang, C. Chu, Y. Zhao.
  The System Description for VoiceMOS Challenge 2022 (KK team, main/ood tasks).

• D. Wang, S. Ye, X. Hu, S. Li.
  The RoyalFlush-NICT System Description for AP21-OLR Challenge (Silk-road team, full tasks).
  In OLR2021 (oriental language recognition challenge), 2021.

• W. Wei, R. Wong, S. Li, Y. Guo and H. Huang.
  System description of Alzheimer's disease early detection (Silk-road team, short speech track).
  In special session of NCMMSC2021 (Alzheimer's disease detection challenge), 2021.

• Y. Han, S. Li, Y. Cao, and M. Yoshikawa,
  System Description for Voice Privacy Challenge (Kyoto Team).
  In special session of INTERSPEECH2020 (VoicePrivacy challenge 2020), 2020.

• Y. Han, Y. Cao, S. Li, Q. Ma, and M. Yoshikawa.
  Voice-Indistinguishability: Protecting Voiceprint with Differential Privacy under an Untrusted Server.
  ACM conference on Computer and Communications Security (CCS), demo, pp. 2125--2127, 2020.

• H. Zhang, S. Li, X. Ma, Y. Zhao, Y. Cao, and T. Kawahara,
  Phantom in the Opera: Effective Adversarial Music Attack on Keyword Spotting Systems.
  in Proc. IEEE-SLT, 2021 (demo session, introduction).

特許（出願・公開/登録）

• 発明者:李勝、ルーシュガン、高島遼一、沈鵬、河井恒・出願人/特許権者：国立研究開発法人情報通信研究機構,
  学習方法・特願2017-236626・特開2019-105899・特許番号「6979203」,
  出願2017年12月11日・公開2019年6月21日・特許登録日2021年11月17日.

• 発明者:高島遼一、李勝、河井恒・出願人/特許権者：国立研究開発法人情報通信研究機構,
  時系列情報の学習システム、方法および ニューラルネットワークモデル・特願2018-044134・特開2019-159654・特許番号「7070894」,
  出願2018年03月12日・公開2018年03月12日・特許登録日2022年5月10日.

• 発明者:李勝、ルーシュガン、高島遼一、沈鵬、河井恒・出願人/特許権者：国立研究開発法人情報通信研究機構,
  音声認識システム、音声認識方法、学習済モデル・特願2018-044491・特開2019-159058・特許番号「7070894」,
  出願 2018年03月12日・公開2018年3月12日, 特許登録日2022年7月22日.

発表・研究会

• K. Soky, S. Li, C. Chu, and T. Kawahara,
  Domain and Language Adaptation of Large-scale Pretrained Model for Speech Recognition of Low-resource Language,
  IEICE-SP, 2022.

• Kai Li, Xugang Lu, Masato Akagi, Jianwu Dang, S. Li, Masashi Unoki,
  Relationship Between Speakers' Physiological Structure and Acoustic Speech Signals:Data-Driven Study Based on Frequency-Wise Attentional Neural Network.
  電子情報通信学会技術研究報告(信学技報), 2022-08. (oral)

• Kak Soky, S. Li, Masato Mimura, Chenhui Chu and Tatsuya Kawahara,
  Leveraging Simultaneous Translation for Enhancing Transcription of Low-resource Language via Cross Attention Mechanism.
  第55回 関西合同音声ゼミ, 2022.

• Hao Shi, Longbiao Wang, S. Li, Jianwu Dang and Tatsuya Kawahara,
  Monaural Speech Enhancement Based on Spectrogram Decomposition for Convolutional Neural Network-sensitive Feature Extraction.
  第55回 関西合同音声ゼミ, 2022.

• X. Chen, H. Huang, and S. Li,
  Adversarial Attack and Defense on Deep Neural Network-based Voice Processing Systems: An Overview.
  in Proc. National Conference on Man-Machine Speech Communication (NCMMSC), 2021.
  (report is selected to publish in Applied Sciences, Special Issues of Machine Speech Communication)

• Y. Han, S. Li, Y. Cao, Q. Ma, and M. Yoshikawa.
  Voice-Indistinguishability — Protecting Voiceprint with Differential Privacy under an Untrusted Server.,
  京都大学 第 15 回 ICT イノベーション, 2021-2-17.

• K. Soky, S. Li, M. Mimura, C. Chu, and T. Kawahara,
  Comparison of End-to-End Models for Joint Speaker and Speech Recognition,
  IEICE-SP, 2021.

• S. Shimizu, C. Chu, S. Li, and S. Kurohashi,
  End-to-End Speech Translation with Cross-lingual Transfer Learning,
  NLP, 2021.

• Y. Han, S. Li, Y. Cao, Q. Ma, and M. Yoshikawa,
  Voice-Indistinguishability: Protecting Voiceprint in Privacy-Preserving Speech Data Release,
  INTERSPEECH 2020 Satellite Workshop (SLIMTS2020) (invited report).

• H. Zhang, S. Ueno, M. Mimura, S. Li, W. Zhang, and T. Kawahara,
  A Mixture of Character and Word End-to-End System for Keyword Spotting,
  INTERSPEECH 2020 Satellite Workshop (SLIMTS2020)(full paper).

• S. Guo, L. Wang, S. Li, J. Zhang, C. Gong, Y. Wang, J. Dang, and K. Honda,
  Investigation of Effectively Synthesizing Code-switched Speech Using Highly Imbalanced Mix-lingual Data and mask embedding,
  INTERSPEECH 2020 Satellite Workshop (SLIMTS2020).

• K. Soky, S. Li, T. Kawahara, and S. Seng,
  Multi-lingual transformer training for Khmer automatic speech recognition,
  INTERSPEECH 2020 Satellite Workshop (SLIMTS2020).

• S. Li, C. Ding, X. Lu, P. Shen, and H. Kawai,
  End-to-End Articulatory Attribute Modeling for Low-resource Multilingual Speech Recognition,
  Acoustical Society of Japan, spring, 2020.

• S. Li, X. Lu, R. Dabre, P. Shen, and H. Kawai,
  Joint Training End-to-End Systems for Speech and Speaker Recognition with Speaker Attributes,
  Acoustical Society of Japan, spring, 2020.

• P. Shen, X. Lu, K. Sugiura, S. Li, H. Kawai,
  Improvement of x-vector for short utterance spoken language identification,
  Acoustical Society of Japan, spring, 2020.

• 李 勝,
  End-to-end音声認識技術の研究,
  情報通信フェア2019, 2019.

• P. Shen, X. Lu, S. Li, and H. Kawai,
  Investigation of multi-domain training for speech recognition,
  Acoustical Society of Japan, spring, 2019.

• 高島 遼一， 李 勝， 河井 恒,
  CTC音響モデルのためのシーケンスレベル知識蒸留法の検討,
  IPSJ SIG-SLP-124-1, 2018.

• S. Li, X. Lu, R.Takashima, P. Shen, and H. Kawai,
  An Empirical Comparison of Sequence Training Methods for the Very Deep Time-delay Neural Network,
  Acoustical Society of Japan, autumn, 2018.

• P. Shen, X. Lu, S. Li, and H. Kawai,
  Short utterance-based spoken language identification,
  Acoustical Society of Japan, autumn, 2018.

• S. Li, X. Lu, R.Takashima, P. Shen, and H. Kawai,
  Improving CTC-based acoustic model with very deep residual neural network,
  Acoustical Society of Japan, spring, 2018.

• R.Takashima, S. Li, and H. Kawai,
  CTC 音響モデルのための knowledge distillation 方式の検討,
  Acoustical Society of Japan, spring, 2018.

• S. Li, X. Lu, P. Shen, and H. Kawai,
  Very deep convolutional residual network acoustic models for Japanese lecture transcription,
  Acoustical Society of Japan, autumn, 2017.

• P. Shen, X. Lu, S. Li, and H. Kawai,
  cGAN-classifier: Conditional Generative Adversarial Nets for Classification,
  Acoustical Society of Japan, autumn, 2017.

• S. Li, X. Lu, S. Sakai, and T. Kawahara,
  Diversity-driven Semi-supervised Ensemble DNN Acoustic Model Training,
  Acoustical Society of Japan, autumn, 2016.

• S. Li, X. Lu, S. Sakai, and T. Kawahara,
  Diversity-driven Semi-supervised Ensemble DNN Acoustic Model Training,
  IPSJ IEICE-SP2016-40, 2016. (oral)

• S. Li, Y. Akita, and T. Kawahara,
  Discriminative data selection from multiple ASR systems' hypotheses for unsupervised acoustic model training,
  IPSJ SIG-SLP-109-8, 2015.(oral)

• S. Li, Y. Akita, T. Kawahara,
  Effective combination of multiple ASR hypotheses with CRF-based classifiers,
  Acoustical Society of Japan, autumn, 2015.

• S. Li, Y. Akita, and T. Kawahara,
  Incorporating divergences from hypotheses of multiple ASR systems to improve unsupervised acoustic model training,
  Acoustical Society of Japan, spring, 2015.

• S. Li, Y. Akita, and T. Kawahara,
  Unsupervised Training of Deep Neural Network Acoustic Models for Lecture Transcriptions,
  Acoustical Society of Japan, autumn, 2014.

• S. Li, Y. Akita, and T. Kawahara,
  Classifier-based data selection for lightly-supervised training of acoustic model for lecture transcription,
  IPSJ SIG-SLP-102-4, 2014.(oral)

• S. Li, Y. Akita, and T. Kawahara,
  Data Selection Assisted by Caption to Improve Acoustic Modeling for Lecture Transcription,
  Acoustical Society of Japan, spring, 2014.(oral)

• S. Li,
  DNN-based Acoustic Modeling and Decoding for Chinese Spontaneous Speech Recogntion,
  Technical-report. 2014.

• S. Li, M. Mimura, and T. Kawahara,
  Automatic Transcription of Chinese Spoken Lectures,
  Acoustical Society of Japan, autumn. 2013.

• S. Li,
  Vocal Tract Length Normalization for Chinese Spontaneous Speech Recogntion,
  Technical-report. 2013.

• S. Li, K. Luo, and L. Wang,
  The Phoneme-level Articulator Dynamics for 3D Pronunciation Animation for Chinese,
  Bulletin of Advanced Technology Research, Vol.5 No.10/Otc.2011, Pages 5-7.

• S. Li, and C. Li,
  Application of the RFID based audio service in regional navigation system,
  Bulletin of Advanced Technology Research, Vol.3 No.2/Feb.2009, Pages 44-47.

ソフトウェア

• VoiceTra: Multilingual Speech Translation Application (our teamwork, strongly suggest trying)

• Julius decoder ── support Kaldi DNN-HMM acoustic model
      └──── support Kaldi feature extractor
      └──── support WFST graph optimization
      └──── support CRF-based confidence estimation
      └──── support EESEN CTC acoustic model

• very deep residual time-delay neural network (TDNN) with LFMMI objective implemented with MS-CNTK

• online speech recognition module for Erica the human robot

データベース

• S.Li, Y.Akita, and T.Kawahara.
  Corpus and transcription system of Chinese Lecture Room.
  In Proc. Int'l Sympo. Chinese Spoken Language Processing (ISCSLP), pp.442--445, 2014.
  (Copy-right belongs to Kawahara Lab. Kyoto Univ.)

• K. Soky, M. Mimura, T. Kawahara, S. Li, C. Ding, C. Chu and S. Sam.
  Khmer Speech Translation Corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC).
  In Proc. O-COCOSDA, 2021.
  (Copy-right belongs to Kawahara Lab. Kyoto Univ., NICT and CADT)

• Kak Soky, Zhuo Gong, S. Li,
  NICT-Tib1: A Public Speech Corpus of Lhasa Dialect for Benchmarking Tibetan Language Speech Recognition Systems.
  in Proc. O-COCOSDA, 2022. (Copy-right belongs to NICT)

• S. Li, J. Li, Q. Liu, Z. Gong,
  Adversarial Speech Generation and Natural Speech Recovery for Speech Content Protection.
  in Proc. LREC (Language Resources and Evaluation Conference), 2022. (Copy-right belongs to NICT)

学会所属, サービス

学会所属

会議運営

査読委員/委員会
ジャーナル

会議

その他

同窓会

Last update: 2022-10-15