M
Mitsuru Nakai
Researcher at Japan Advanced Institute of Science and Technology
Publications - 32
Citations - 692
Mitsuru Nakai is an academic researcher from Japan Advanced Institute of Science and Technology. The author has contributed to research in topics: Phrase & Handwriting recognition. The author has an hindex of 11, co-authored 32 publications receiving 686 citations. Previous affiliations of Mitsuru Nakai include Tohoku University & Toyama Prefectural University.
Papers
More filters
Proceedings Article
Dynamic Time-Alignment Kernel in Support Vector Machine
TL;DR: The proposed SVM (DTAK-SVM) is evaluated in speaker-dependent speech recognition experiments of hand-segmented phoneme recognition and preliminary experimental results show comparable recognition performance with hidden Markov models (HMMs).
Proceedings ArticleDOI
Substroke approach to HMM-based on-line Kanji handwriting recognition
TL;DR: A new method is proposed for online handwriting recognition of Kanji characters that employs substroke HMM as minimum units to constitute Japanese KanjiCharacters and utilizes the direction of pen motion to fully utilize the continuous speech recognition algorithm.
Proceedings ArticleDOI
Multiple-regression hidden Markov model
TL;DR: Experiments of speaker-dependent isolated word recognition demonstrated that MR-HMMs using F/sub 0/ based auxiliary features reduced the error rates by more than 20% compared with the conventional HMMs.
Proceedings Article
Support vector machine with dynamic time-alignment kernel for speech recognition.
Proceedings ArticleDOI
Context-dependent substroke model for HMM-based on-line handwriting recognition
TL;DR: The substroke HMM approach where a modeling unit "substroke" that is much smaller than a whole character is employed and each character is modeled as a concatenation of only 25 kinds of substrok HMMs is proposed.