H
Hiroshi Fujimura
Researcher at Toshiba
Publications - 42
Citations - 169
Hiroshi Fujimura is an academic researcher from Toshiba. The author has contributed to research in topics: Acoustic model & Sentence. The author has an hindex of 6, co-authored 40 publications receiving 157 citations. Previous affiliations of Hiroshi Fujimura include Nagoya University.
Papers
More filters
Patent
Electronic device and control method
TL;DR: In this article, an electronic device includes a touch panel, an operation module, a detector, and a switch module, and the detector detects whether the touch panel is wet based on a touch position detected by the touch panels.
Journal ArticleDOI
Construction and Evaluation of a Large In-Car Speech Corpus
Kazuya Takeda,Hiroshi Fujimura,Katsunobu Itou,Nobuo Kawaguchi,Shigeki Matsubara,Fumitada Itakura +5 more
TL;DR: A system specially built into a Data Collection Vehicle which supports the synchronous recording of multichannel audio data from 16 microphones that can be placed in flexible positions, multich channel video data from 3 cameras, and vehicle related data is developed.
Proceedings ArticleDOI
The Toshiba entry to the CHiME 2018 Challenge
Rama Doddipatla,Takehiko Kagoshima,Cong-Thanh Do,Petko N. Petkov,Catalin-Tudor Zorila,Euihyun Kim,Daichi Hayakawa,Hiroshi Fujimura,Yannis Stylianou +8 more
TL;DR: This paper summarises the Toshiba entry to the single-array track of the CHiME 2018 speech recognition challenge, based on conventional acoustic modelling (AM), where phonetic targets are tied to features at the frame-level, and use the provided tri-gram language model.
Patent
Apparatus and Method for Speech Recognition
Hiroshi Fujimura,Takashi Masuko +1 more
TL;DR: In this paper, a speech recognition apparatus includes a first storing unit configured to store a first acoustic model invariable regardless of speaker and environment, a second storing unit configurable to classify second acoustic models, and a calculating unit to calculate a second likelihood for each of the groups with regard to the input speech by use of the calculation result on the shared parameters and the non-shared parameters of the classification model.
Patent
Speech recognition apparatus and method and program therefor
TL;DR: In this paper, a speech recognition system includes a generating unit generating a speech-feature vector expressing a feature for each of frames obtained by dividing an input speech, a storage unit storing a first acoustic model obtained by modeling a feature of each word by using a state transition model, a calculation unit calculating a second probability of transition to the at-end-frame state on the corresponding-transition-path, and a finding unit finding to which word the input speech corresponds based on the maximum probability and the second probability.