
Hiromitsu Nishizaki
西﨑 博光
Professor, Faculty of Engineering, University of Yamanashi
Chair, Department of Mechanical Engineering (Head, Mechatronics Course)
Director, Center for Promotion of Internationalization
Head of the Nishizaki-Leow Lab (ALPS Lab)
AI that understands people and works alongside them,
in classrooms, on farms and in everyday life.
Paper counts from the lab publication database (November 2026); citation metrics from Google Scholar (checked November 16, 2026).
About
経歴Born in Sayo, Hyogo. After Tsuyama National College of Technology, I transferred to Toyohashi University of Technology and received my Dr. Eng. in 2003. I joined the University of Yamanashi the same year and became a full professor in 2022 after serving as research associate, assistant professor and associate professor. In 2015–2016 I was a visiting researcher at National Taiwan University. Speech research at Yamanashi began with Dr. Minoru Shigenaga in 1955 and continued under Dr. Yoshihiro Sekiguchi; I have led the lab since 2013. My own research began with speech recognition and spoken document retrieval, expanded to image recognition and sensor-based sensing, and now focuses on physical AI that works in the real world.
Career
- Chair, Department of Mechanical Engineering (Head, Mechatronics Course), University of Yamanashi
- Director, Center for Promotion of Internationalization, University of Yamanashi
- Professor, Graduate Faculty of Interdisciplinary Research, University of Yamanashi
- Associate Professor, University of Yamanashi
- Visiting Researcher, College of EECS, National Taiwan University
- Assistant Professor, University of Yamanashi
- Research Associate, University of Yamanashi
Education
- Dr. Eng., Toyohashi University of TechnologyThesis on spoken-query information retrieval for spoken documents
- M.Eng., Toyohashi University of Technology
- B.Eng., Toyohashi University of Technology
- Associate Degree in Engineering, Tsuyama National College of Technology (Information Engineering)
Memberships
Research
研究I work on spoken language processing and multimedia pattern recognition. Starting from deep-learning-based speech recognition and natural language processing, my lab's work extends to LLM-based dialogue and education support, handwritten character recognition, and smart agriculture and physical AI (farm robots) for Yamanashi's Shine Muscat grapes. I have led the lab since 2013.
Speech recognition
Large-vocabulary continuous speech recognition, combining multiple recognizers, recognition of lecture, elderly and multilingual speech, lightweight end-to-end models for embedded devices, speech-to-braille and captioning.
143 papers & talksSpeech & audio signal processing
Speech enhancement for hearing aids, sound event and acoustic scene classification, speaker recognition and anti-spoofing, speech-based depression detection, waveform-based acoustic features.
68 papers & talksSpoken document retrieval
Spoken term detection and spoken document retrieval, including organizing and taking part in NTCIR evaluation tasks and web-based document expansion for indexing.
105 papers & talksNLP & dialogue
Applications of large language models, spoken dialogue systems and dialogue robots, paraphrasing and translation, text classification and topic estimation.
52 papers & talksEducation & accessibility
Lecture speech evaluation and speaking training, classroom analysis, information access for deaf and hard-of-hearing people, note-taking support, customer-service training and language-barrier-free classes.
70 papers & talksCharacter & image recognition
Handwritten character generation and recognition, document OCR with vision-language models, similar-drawing search, face and pose estimation from fisheye images, 3D reconstruction.
28 papers & talksSmart agriculture & physical AI
Centered on Yamanashi's Shine Muscat grapes: color and taste estimation from images, the berry-thinning app tsubura, and produce sorting, extending to physical AI that acts in the real world, such as harvesting, thinning and autonomous mobile robots using camera/LiDAR perception and multimodal LLMs.
49 papers & talksHealthcare & sensing
Applications to healthcare and IoT: medical record and patient information systems, ALS diagnosis support, pose-based gait and motion assessment, swallowing assessment, speech-based depression detection, UWB-radar breathing monitoring, and air-quality and smart-home sensing.
51 papers & talksJSPS KAKENHI (as Principal Investigator)
- Realizing language-barrier-free classes with artificial intelligence
- Information assurance and class improvement support technology through advanced nonverbal modality sensing in classes
- Building and evaluating spoken dialogue system technology for customer-service training
- A technology transfer system based on creation of work records and procedures using speech and language processing technologies
- Study on an electronic note-taking support system using speech and language processing technologies
- Improvement of spoken term detection and its application to speech recognition and spoken document retrieval
- Training system for improving the manner of speaking in lectures using speech processing technologies
- Study on construction of university lecture speech content and its evaluation
Publications
業績At a glance
Publications & talks
External funding
Google Scholar
Sources: lab publication database (papers, presentations, talks), KAKEN (KAKENHI), researchmap (other grants), Google Scholar (citations, checked November 16, 2026). Internationally co-authored papers are journal and conference papers with at least one author at a non-Japanese institution (affiliations checked on OpenAlex). Other figures as of November 2026.
Papers per fiscal year
* Japanese fiscal years run from April to March. FY2026 is in progress.
Show as a table
| FY | Journal | Int'l conference |
|---|---|---|
| FY2026 | 5 | 11 |
| FY2025 | 7 | 10 |
| FY2024 | 7 | 12 |
| FY2023 | 10 | 9 |
| FY2022 | 7 | 7 |
| FY2021 | 4 | 13 |
| FY2020 | 2 | 12 |
| FY2019 | 1 | 11 |
| FY2018 | 1 | 3 |
| FY2017 | 0 | 5 |
| FY2016 | 1 | 4 |
| FY2015 | 1 | 2 |
| FY2014 | 1 | 4 |
| FY2013 | 2 | 4 |
| FY2012 | 2 | 3 |
| FY2011 | 1 | 7 |
| FY2010 | 2 | 5 |
| FY2009 | 0 | 2 |
| FY2008 | 2 | 4 |
| FY2007 | 0 | 1 |
| FY2006 | 2 | 0 |
| FY2005 | 2 | 0 |
| FY2004 | 5 | 3 |
| FY2003 | 4 | 3 |
Citations per year (Google Scholar)
* 2026 is year-to-date (checked November 16, 2026)
Show as a table
| Year | Citations |
|---|---|
| 2026 | 186 |
| 2025 | 235 |
| 2024 | 204 |
| 2023 | 171 |
| 2022 | 153 |
| 2021 | 90 |
| 2020 | 53 |
| 2019 | 40 |
| 2018 | 21 |
| 2017 | 52 |
| 2016 | 72 |
| 2015 | 47 |
| 2014 | 96 |
| 2013 | 120 |
| 2012 | 43 |
| 2011 | 56 |
| 2010 | 36 |
| 2009 | 28 |
| 2008 | 23 |
| 2007 | 7 |
| 2006 | 18 |
| 2005 | 21 |
| 2004 | 22 |
| 2003 | 27 |
| 2002 | 7 |
Recent journal papers (latest 10)
- 2027Dual-range vision-based cutting-point localization and end-effector positioning for robotic table grape inflorescence trimmingPanitan Muangkammuen, Prawit Buayai, Sopheapanha Bun, Sonthaya Buayai, Xiaoyang Mao, Hiromitsu NishizakiComputers and Electronics in Agriculture, vol. 257, Art. no. 112651
- 2026.09A dual-stream spectral network for versatile full-band speech enhancementJunkang Yang, Hiromitsu Nishizaki, Chee Siang Leow, Hongqing Liu, Lu GanThe Journal of the Acoustical Society of America, vol. 160, no. 3, pp. 2272–2287
- 2026.09Lightweight streaming speech enhancement for AIoT-enabled wearable hearing aids using parallel spiking MambaJunkang Yang, Hiromitsu Nishizaki, Chee Siang Leow, Hongqing Liu, Lu GanInternet of Things, vol. 39, Art. no. 102046
- 2026.09Adversarial audio inpainting with selective state space model, efficient attention, and large-scale pre-trained modelJunkang Yang, Hongqing Liu, Liming Shi, Lu Gan, Hiromitsu Nishizaki, Chee Siang LeowArchives of Acoustics, vol. 51, no. 3, pp. 515–532
- 2026.04End-to-End Position Prediction for Robotic Grape Berry ThinningYin Suan Tan, Prawit Buayai, Dear Moeurn, Hiromitsu Nishizaki, Koji Makino, Xiaoyang MaoJournal of Robotics and Mechatronics, Vol.38, No.2, pp.513-524
- 2025.12Enhancing the robustness of speech anti-spoofing countermeasures through joint optimization and transfer learningYikang Wang, Xingming Wang, Chee Siang Leow, Qishan Zhang, Ming Li, Hiromitsu NishizakiIEICE Transactions on Information and Systems, vol. E108-D, no. 12, pp. 1594–1604
- 2025.11ColabReaction: Accelerating transition state searches with machine learning potentials on Google ColaboratoryMasayuki Karasawa, Chee Siang Leow, Hideaki Yajima, Shuta Arai, Hiromitsu Nishizaki, Tohru Terada, Hajime SatoJournal of Chemical Information and Modeling, vol. 65, no. 21, pp. 11908–11914
- 2025.11Non-invasive estimation of Shine Muscat grape color and sensory evaluation from standard camera imagesRyosuke Shimazu, Chee Siang Leow, Prawit Buayai, Xiaoyang Mao, Wan-Young Chung, Hiromitsu NishizakiThe Visual Computer, vol. 41, no. 14, pp. 11705–11720
- 2025.09Optimizing mRNA Vaccine Degradation Prediction via Penalized Dropout ApproachesHwai Ing Soon, Azian Azamimi Abdullah, Hiromitsu Nishizaki, Latifah Munirah KamarudinIEEE Access, Vol. 13, pp. 137561-137578, ISSN: 2169-3536
- 2025.08Handwritten character image generation for effective data augmentationChee Siang Leow, Tomoki Kitagawa, Hideaki Yajima, Hiromitsu NishizakiIEICE Transactions on Information and Systems, vol. E108-D, no. 8, pp. 967–976
IF = the journal's latest Journal Impact Factor (Clarivate Journal Citation Reports; 2025 value, or the previous year's value with the year shown). SJR = SCImago Journal Rank (2025); Q1–Q4 = the journal's best SCImago quartile across its subject categories (hover to see the category). None of these are values at the time of publication.
Recent conference papers (latest 10)
- 2027.01DMPNet: Versatile Speech Enhancement Based on Differential Mixing and Bidirectional Spectral PropagationJunkang Yang, Hiromitsu Nishizaki, Chee Siang Leow, Hongqing Liu, Lu GanProceedings of the 33rd International Conference on MultiMedia Modeling (MMM 2027), to appear
- 2026.12Portable Lemon Grading System Featuring a Projection-Guided Free-Placement WorkspaceYe-Seul Jang, Prawit Buayai, Panitan Muangkammuen, Suit Mun Ng, Hiromitsu Nishizaki, Xiaoyang MaoProceedings of the 2026 International Conference on Cyberworlds (CW 2026), to appear
- 2026.10A Harmonic Denoising Approach for Non-Contact Breathing Rate Estimation Using Ultra-Wideband RadarSiti Mahfuzah Fauzi, Latifah Munirah Kamarudin, Intan Kartika Kamarudin, Hiromitsu Nishizaki, Chee Siang Leow, Ammar ZakariaProceedings of the 2026 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2026)
- 2026.10Benchmarking and Validation of Pre-Trained 3D Pose Estimation Models for Sports Action ClassificationThenmoli Manogar, Ammar Zakaria, Ahmad Shakaff Ali Yeon, Shuhaida Yahud, Syed Muhammad Mamduh Syed Zakaria, Latifah Munirah Kamarudin, Hiromitsu Nishizaki, Chee Siang LeowProceedings of the 2026 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2026)
- 2026.10Fisheye Geometry-Aware Egocentric 3D Human Pose Estimation with Patch Viewing DirectionsYinghao He, Chee Siang Leow, Hiromitsu NishizakiProceedings of the 2026 IEEE 15th Global Conference on Consumer Electronics (GCCE 2026)
- 2026.10AI-Driven Continuous Optimization of Particle Damper Fill Ratio for Hand-Transmitted Vibration Mitigation in Handheld Grass-Cutting MachineryMuhammad Aqiff Bin Masri, Muhajir Ab Rahim, Latifah Munirah Kamarudin, Hiromitsu Nishizaki, Ammar Zakaria, Syed Muhammad Mamduh Syed Zakaria, Muhamad Norhisham Abdul Rani, Mohd Azmi Yunus, Zheng ChewProceedings of the 2026 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2026)
- 2026.10Performance Comparison of Multi-Layer Perceptron and Deep Multi-Layer Perceptron for RSSI-Based Device-Free LocalizationShaufikah Shukri, Latifah Munirah Kamarudin, Hiromitsu NishizakiProceedings of the 2026 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS 2026)
- 2026.09Harmonic and entropy neural modeling with demographic fusion for speech-based depression detectionJunkang Yang, Hiromitsu Nishizaki, Chee Siang Leow, Lu Gan, Hongqing LiuProceedings of the 35th International Conference on Artificial Neural Networks (ICANN 2026), Lecture Notes in Computer Science, Vol. 17092, pp. 623–634
- 2026.09Difficulty-Aware Curriculum and Token-Weighted Rationale Distillation from LLM for Sequential RecommendationJitong Jiang, Dongjin Yu, Changgeng Li, Dongjing Wang, Hiromitsu Nishizaki, Chee Siang LeowProceedings of the 2026 Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML PKDD 2026), pp. 177–194
- 2026.08Reference-free handwritten Japanese character generation via CLIP-conditioned diffusion modelsKoki Fujita, Hideaki Yajima, Chee Siang Leow, Hiromitsu NishizakiProceedings of the 2026 International Conference on Document Analysis and Recognition (ICDAR 2026), pp. 3–19
Books (in Japanese)
- 2022.10Proceedings of the Dialogue Robot Competition 2022Ryuichiro Higashinaka, Takashi Minato, Hiromitsu Nishizaki, Takayuki Nagai(編)公式予稿集(arXiv:2210.12034)
- 2022.07あそべる! 通じ合う! てづくりAIロボット牧野浩二, 和田義久, 西崎博光 ほか誠文堂新光社
- 2021.06たのしくできる 深層学習&深層強化学習による電子工作 TensorFlow編牧野浩二, 西崎博光東京電機大学出版局
- 2021.02TensorFlowによる深層強化学習入門 — OpenAI Gym + PyBulletによるシミュレーション牧野浩二, 西崎博光オーム社
- 2020.03たのしくできる深層学習牧野浩二, 西崎博光東京電機大学出版局
- 2019実践!Chainerとロボットで学ぶディープラーニング丸山史郎, 牧野浩二, 西﨑博光アフレル
- 2018.08Pythonによる深層強化学習入門 — ChainerとOpenAI Gymではじめる強化学習牧野浩二, 西崎博光オーム社
- 2018.03人工知能を作る小池誠, 鎌田智也, 足立悠, 西崎博光, 牧野浩二 ほか(16名)CQ出版社
- 2018.02算数&ラズパイから始めるディープ・ラーニング牧野浩二, 西崎博光CQ出版社
Magazine articles (in Japanese)
- 2025.10ColabReaction:反応の核心に、誰でも飛び込める時代へ佐藤玄, 唐澤昌之, 寺田透, レオ・チー・シャン, 矢島英明, 荒井秀太, 西﨑博光酵素工学ニュース(酵素工学研究会誌), 第94号, pp.33-37, 酵素工学研究会
- 2024.01シャインマスカットの収穫時期をAIで判断牧野浩二, 西崎博光, Leow Chee Siang, Prawit Buayai, 茅暁陽Interface 2024年1月号, pp.218-219, CQ出版社
- 2023.09これからの自然言語処理西崎博光これからのコンピュータ技術555(第1部第2章人工知能), インターフェース(Interface), pp.52-56, CQ出版社
- 2022.03時系列データを分析・分類するための深層学習技術西崎博光化学工業 Vol.73, No.3, pp.167-173
- 2021.05音声認識の仕組みおよび現状と将来展望西崎博光月刊電気計算 2021年5月号, 電気書院
- 2021スマートグラスを用いたブドウ収穫支援システム西崎博光日本機械学会関東支部ニュースレター No.51, p.11
- 2020.02人工知能アルゴリズム探検隊 第35回 深層強化学習牧野浩二, 西崎博光Interface 2020年2月号, pp.156-160, CQ出版社
- 2020.02ディープラーニングの2大フレームワーク牧野浩二, 西崎博光Interface 2020年2月号, pp.22-23, CQ出版社
- 2020.02これまでの/これからの自然言語処理16+α西崎博光Interface 2020年2月号, pp.31-36, CQ出版社
- 2020.01人工知能アルゴリズム探検隊 第34回 AI走行ロボに別の学習データを追加で教える牧野浩二, 西崎博光Interface 2020年1月号, pp.144-152, CQ出版社
- 2018.12My人工知能の育て方 第1部第3章 My「音」認識人工知能のうまい育て方牧野浩二, 西崎博光, 澤田直輝Interface 2018年12月号, pp.38-55, CQ出版社
- 2017.12第3章 注目ディープ・ラーニング用ライブラリ百科牧野浩二, 西崎博光Interface 2017年12月号,pp.24--32,CQ出版社,2017年10月25日発売
- 2017.09人工知能アルゴリズム探検隊 第11回番外編 インテル人工知能向け小型コンピュータの実力チェック牧野浩二, 西崎博光Interface 2017年9月号,pp.130--134, CQ出版社,2017年7月25日発売
- 2017.08算数&ラズパイから始める人気AI ディープ・ラーニング牧野浩二, 西崎博光Interface 2017年8月号,pp.17--127, CQ出版社,2017年6月24日発売
- 2017.033大人工知能ライブラリ+α牧野浩二, 西崎博光Interface 2017年3月号, pp.87-91, CQ出版社
Patents (Japan)
- 2025.11農作物処理システムレオ チーシャン, 堀川萌樹, 鈴木裕真, 西崎博光特願2025-199336, 2025年11月19日出願
- 2022.04色判定システム、色判定方法および、色判定プログラム西崎博光, 花田隆貴特許第7064201号, 2022年4月26日登録(特願2021-146701, 2021年9月9日出願)
- 2019.02信号変換システム、特徴出力システム、機械学習システムおよび信号変換プログラム西崎博光, 澤田直輝特願2019-020267, 2019年2月出願
- 2015.03言語モデル・音声認識辞書作成装置及びそれらにより作成された言語モデル・音声認識辞書を用いた情報処理装置西崎博光, 上平拓弥, 高岡充, 関口芳廣特許第5703491号, 2015年3月登録(特願2010-014372, 2010年1月26日出願)
- 2014.05電子ノート作成支援装置および電子ノート作成支援装置用プログラム西崎博光, 太田晃平, 関口芳廣特開2014-085998, 2014年5月12日公開(特願2012-236445, 2012年10月26日出願)
Awards
- Excellent Poster Award (Gold Prize), IEEE GCCE 2026 (co-author)
- Best Paper Award, ICANN 2026
- Excellent Presentation Award, SICE (co-author)
- IEICE Educational Contribution Award
- Best Paper Award, Cyberworlds 2024
- Outstanding Faculty Award, University of Yamanashi
- Outstanding Faculty Award, University of Yamanashi
- Excellent Presentation Award, SICE SI2021
- IPSJ Excellent Teaching Material Award
- Excellent Paper Award (Outstanding Prize), IEEE GCCE 2020
- Best Paper Award, O-COCOSDA 2019
- Haneda Robotics Lab Award, SoftBank
Teaching
授業Courses in AY2026. Course names link to the university syllabus (in Japanese).
Undergraduate (Mechatronics)
- Multimedia EngineeringSpring, 3rd yr
- System DesignSpring, 3rd yr
- Numerical ComputationFall, 3rd yr
- Embedded Programming I (Exercises)Spring, 2nd yr
- Embedded Programming IIFall, 2nd yr
University-wide
Graduate
- Advanced Communication and Control NetworksMaster's, Spring
- Advanced Man-Machine InterfacesDoctoral, Fall
Abroad
- Deep Learning for Multimedia (Hangzhou Dianzi University, China)Intensive
Contact
連絡先- Affiliation
- Faculty of Engineering, University of Yamanashi
- Teaching
- Undergraduate: Mechatronics Course, Faculty of Engineering
Master's: Mechatronics Engineering Course
Doctoral: System Integration Engineering Course - Address
- Room A1-313, A-1 Building 3F, Faculty of Engineering
4-3-11 Takeda, Kofu, Yamanashi 400-8511, Japan - Access
- About 15 min on foot or 5 min by bus from JR Kofu Station
- hnishi [at] yamanashi.ac.jp
- Lab
- www.alps-lab.org
- X
- @hiro32_n
Researcher IDs
- Google ScholarUGpdyLMAAAAJ
- researchmapnishizaki
- ORCID0000-0002-7717-8312
- KAKEN researcher no.40362082
- DBLP87/1050







