K
Kazunori Komatani
Researcher at Osaka University
Publications - 237
Citations - 3192
Kazunori Komatani is an academic researcher from Osaka University. The author has contributed to research in topics: Humanoid robot & Recurrent neural network. The author has an hindex of 28, co-authored 226 publications receiving 2981 citations. Previous affiliations of Kazunori Komatani include Nagoya University & Kyoto University.
Papers
More filters
Proceedings Article
Hybrid collaborative and content-based music recommendation using probabilistic model with latent user preferences
TL;DR: A hybrid music recommendation method that solves problems of two prominent conventional methods: collaborative filtering and content-based recommendation and can reasonably recommend pieces even if they have no ratings is presented.
Proceedings ArticleDOI
Sound source localization based on deep neural networks with directional activate function exploiting phase information
Ryu Takeda,Kazunori Komatani +1 more
TL;DR: This paper describes sound source localization (SSL) based on deep neural networks (DNNs) using discriminative training and indicates that the method outperformed the naive DNN-based SSL by 20 points in terms of the block-level accuracy.
Journal ArticleDOI
An Efficient Hybrid Music Recommender System Using an Incrementally Trainable Probabilistic Generative Model
TL;DR: A probabilistic generative model is used that unifies the collaborative and content-based data in a principled way that accurately recommended pieces including nonrated ones from a wide variety of artists and maintained a high degree of accuracy even when new users and rating scores were added.
Proceedings ArticleDOI
Discriminative multiple sound source localization based on deep neural networks using independent location model
Ryu Takeda,Kazunori Komatani +1 more
TL;DR: The experiments indicated that the SSL based on DNNs trained by the proposed training method out-performed a conventional SSL method by a maximum of 18 points in terms of block-level correctness.
Journal ArticleDOI
Instrument identification in polyphonic music: feature weighting to minimize influence of sound overlaps
TL;DR: A new solution to the problem of feature variations caused by the overlapping of sounds in instrument identification in polyphonic music by weighting features based on how much they are affected by overlapping, which improves instrument identification using musical context.