K
Kazuya Takeda
Researcher at Nagoya University
Publications - 546
Citations - 9667
Kazuya Takeda is an academic researcher from Nagoya University. The author has contributed to research in topics: Speech processing & Speech enhancement. The author has an hindex of 42, co-authored 495 publications receiving 7719 citations. Previous affiliations of Kazuya Takeda include Kobe Women's University & Nara Institute of Science and Technology.
Papers
More filters
Proceedings ArticleDOI
Two-stage noise spectra estimation and regression based in-car speech recognition using single distant microphone
TL;DR: Compared to the original noisy speech, the proposed regression-based approach obtains an average relative word error rate (WER) reduction of 65% in the authors' isolated word recognition experiments conducted in 12 real car environments.
Proceedings ArticleDOI
Feature transformation based on discriminant analysis preserving local structure for speech recognition
TL;DR: Two methods for locality preserving HDA and locality preserving PLDA are introduced and an efficient calculation scheme is given to obtain an optimal projection.
Posted Content
Road Scene Graph: A Semantic Graph-Based Scene Representation Dataset for Intelligent Vehicles.
TL;DR: This paper proposes road scene graph, a special scene-graph for intelligent vehicles, which provides not only object proposals but also their pair-wise relationships in a topological graph.
Proceedings ArticleDOI
Spectral weighting of SBCOR for noise robust speech recognition
TL;DR: It is derived that SBCOR results in the lateral inhibitive weighting (LIW) processing of the power spectrum, and it is shown that the LIW is significantly effective for noise robust acoustic analysis using a DTW word recognizer.
Book ChapterDOI
A Stochastic Approach for Modeling Lane-Change Trajectories
TL;DR: A signal-processing approach for modeling vehicle trajectory during lane changes while driving with a hidden Markov model and a cognitive distance space represented with a hazard-map function is discussed.