K
Keiichi Osako
Researcher at Sony Broadcast & Professional Research Laboratories
Publications - 29
Citations - 259
Keiichi Osako is an academic researcher from Sony Broadcast & Professional Research Laboratories. The author has contributed to research in topics: Audio signal & Signal. The author has an hindex of 6, co-authored 29 publications receiving 247 citations. Previous affiliations of Keiichi Osako include Nara Institute of Science and Technology & Carnegie Mellon University.
Papers
More filters
Journal ArticleDOI
Blind Spatial Subtraction Array for Speech Enhancement in Noisy Environment
TL;DR: It is theoretically and experimentally pointed out that ICA is proficient in noise estimation under a non-point-source noise condition rather than in speech estimation, and a new blind spatial subtraction array (BSSA) is proposed that utilizes ICA as a noise estimator.
Proceedings ArticleDOI
Complex recurrent neural networks for denoising speech signals
TL;DR: Noise reduction experiments on noisy speech, both with digitally added synthetic noise and real car noise, show that the proposed algorithm can recover much of the degradation caused by the noise.
Patent
Mastication detection device and mastication detection method
TL;DR: In this paper, an information processing apparatus and method provide logic for processing information, including a receiving unit configured to receive an audio signal associated with a motion of a human mandible over a time period.
Patent
Audio signal processing device, audio signal processing method, and program
TL;DR: In this paper, an audio signal processing device includes a first microphone configured to pick up audio and output a first audio signal, a second microphone configured for picking up the audio signal and outputting a second audio signal; and an operating sound reducing unit configured to reduce the estimated operating sound spectrum signal.
Proceedings ArticleDOI
Supervised monaural source separation based on autoencoders
TL;DR: A new supervised monaural source separation based on autoencoders using the autoencoder for the dictionary training such that the nonlinear network can encode the target source with high expressiveness.