日本語
The University of Electro-Communications 
Graduate School of Informatics and Engineering, School of Informatics and Engineering 
Department of Computer and Network Engineering, Cluster I (Informatics and Computer Engineering) 

Assistant Professor 
Toru NAKASHIKA 

Personal website  
Personal website  

Career
INSA Lyon  LIRIS, CNRS  Researcher  2011/09/01-2012/08/25 
Kobe University  Graduate School of System Informatics  Assistant Professor  2014/10/01-2015/03/31 

Academic background
Kobe University  Faculty of Engineering  Department of Computer Science and Systems Engineering  Other  2009/03/25  Graduated 
Kobe University  Graduate School of Engineering  Department of Computer Science and Systems Engineering  Master's program  2011/03/25  Completed 
Kobe University  Graduate School of System Informatics  Department of Information Science  Doctoral program  2014/09/25  Graduated 

Academic degrees
Ph.D. in Engineering  Kobe University  2014/09/25 

Classes responsible within the university
2017  コンピュータサイエンス実験第一  Spring semester  Undergraduate, daytime 
View details...

Current research areas
Multimedia database 
Cognitive science 
Perceptual information processing 

Research keywords
Multimedia Signal Processing 
Voice Conversion 
Deep Learning 

Current research subjects
Simultaneous Emotion/Phoneme/Identity Separation and Intention Estimation  2015/04/01-Present 
Discovery of Mechanism for Human Activity and Recognition Based on Statistical Modelling  2015/04/01-Present 

Published papers
Paper  Yes  Speaker-adaptive-trainable Boltzmann machine and its application to non-parallel voice conversion  Joint  Toru Nakashika, Yasuhiro Minami  EURASIP Journal on Audio, Speech, and Music Processing  DOI: 10.1186/s13636-017-0112-6, 1-10  2017/06/29 
Paper  Yes  Non-Parallel Training in Voice Conversion Using an Adaptive Restricted Boltzmann Machine  Joint  Toru Nakashika, Yasuhiro Minami  IEEE/ACM Transactions on Audio, Speech and Language Processing  24/ 11, 2032-2045  2016/08 
Paper  Yes  Phone Labeling Based on the Probabilistic Representation for Dysarthric Speech Recognition  Joint  Yuki Takashima, Toru Nakashika, Tetsuya Takiguchi, and Yasuo Ariki  American Journal of Signal Processing  6/ 1, 19-23  2016/06 
Paper  Yes  Small-parallel exemplar-based voice conversion in noisy environments using affine non-negative matrix factorization  Joint  Ryo Aihara, Takao Fujii, Toru Nakashika, Tetsuya Takiguchi, Yasuo Ariki  EURASIP Journal on Audio, Speech, and Music Processing  2015:32/ DOI: 10.1186/s13636-015-0075-4, 1-9  2015/11/25 
Paper  Yes  Voice Conversion Using RNN Pre-Trained by Recurrent Temporal Restricted Boltzmann Machines  Joint  Toru Nakashika, Tetsuya Takiguchi, and Yasuo Ariki  IEEE/ACM Transactions on Audio, Speech and Language Processing  23/ 3, 580-587  2015/03 
View details...

Research presentations
Oral presentation for an academic conference  GCDRMを用いたテキスト・音声の同時確率表現に基づく音声認識・合成器の同時構築  日本音響学会2017年秋季研究発表会  1-8-2  No  曾根健太郎, 中鹿亘  2017/09 
Oral presentation for an academic conference  複素 RBM を用いた音声スペクトルモデリングの改良と評価  日本音響学会2017年秋季研究発表会  1-8-1  No  中鹿亘, 高木信二, 山岸順一  2017/09 
Oral presentation for an academic conference  フェイクデータを用いた ARBM に基づく非パラレル声質変換手法の改善  日本音響学会2017年秋季研究発表会  2-Q-28  No  中鹿亘  2017/09 
Oral presentation for an academic conference  Practice Process Analysis Using Score Matching Method Based on OBE-DTW and Its Effects on Memorizing Musical Score  International Symposium on Performance Science (ISPS) 2017  No  Toru Nakashika and Eriko Aiba  2017/08 
Oral presentation for an academic conference  複素RBM:制限ボルツマンマシンの複素数拡張と音声信号への応用と評価  情報処理学会音声言語研究会技術研究報告  2017-SLP-117/ 9  No  中鹿亘,高木信二,山岸順一  2017/07 
View details...

Intellectual property rights
Patent  声質変換装置、声質変換方法およびプログラム  特願2017-036109  2017/02/28  Filed 
Patent  符号化装置、符号化方法およびプログラム  特願2017-037640  2017/02/28  Filed 
Patent  声質変換装置、声質変換方法および声質変換プログラム  特願2016-032488  2016/02/23  Filed 
Patent  声質変換方法および声質変換装置  特願2015-114238  2015/06/04  Filed 
View details...

Awards
SIGMUS Excellent Presentation Award  Speech Modeling Using Three-Way Restricted Boltzmann Machine for Simultaneous Speaker-Phoneme Recognition  2016/05/22 
IEICE ISS Young Researcher's Award in Speech Field  Speaker-Dependent Conditional Restricted Boltzmann Machine for Voice Conversion  2014/06/20 
Student Award  Voice Conversion Based on Non-negative Matrix Factorization Using Phoneme Categorized Dictionary  2014/03/11 
View details...

External fund
制限ボルツマンマシンの複素数拡張と音声合成への応用  Donations for research  電気通信普及財団  2017/04/01-Present 
View details...

Past records of joint or funded research
複素数表現に着目した確率モデルに基づく声質変換技術を用いた構音障がい者発話支援  2017-Present  Joint research  Domestic joint research 
エネルギー関数に基づく非パラレル声質変換技術を用いた軽度構音障がい者の発話支援に関する研究  2016-Present  Joint research  Domestic joint research 
View details...

Memberships of academic societies
電子情報通信学会  2016/04/01-Present 
IEEE Membership  2016/01/01-Present 
日本音響学会  2015/04/01-Present 
View details...

Industry-academia-government collaboration activities
Industry-academia-government collaboration  JST主催イノベーションジャパン2016  実用的なボシスチェンジャー  2016/08-Present 
Industry-academia-government collaboration  JST主催新技術説明会  学習が”容易”な声質変換技術  2016/06-Present 
Industry-academia-government collaboration  TAMA協会(首都圏産業活性化協会)  ソリューション提案交流会  2016/02-Present 
View details...

Media coverage
Radio  TBSラジオ  THE FROGMAN SHOW A.I.共存ラジオ 好奇心家族  2017/11/10  20:00~20:30 
Television  フジテレビジョン  まさか!それがデキるのか!?世界最高技術でやっちゃいましたSP  2017/07/15 
Television  フジテレビジョン  ホンマでっか!?TV  2017/05/24 
Other  Nikkei Business Publications, Inc.  Nikkei Electronics 2016 August  2016/07/20 
Newspaper  Nikkei Inc.  The Nikkei  2016/05/16 
View details...

Committee activity in the University
卒研担当  2017/04-Present 
FS専攻教務委員  2017/04-Present 
FS専攻教務委員  2016/04-Present 
技術英語実施委員会委員  2016/04-2017/03 
FS専攻教務委員  2015/04-Present 
View details...