Atsuhiko Kai

Learn More
1 Japan Advanced Institute of Science and Technology 2 The University of Tokyo 3 Toyohashi University of Technology 4 Advanced Telecommunications Research Institute International 5 National Institute of Advanced Industrial Science and Technology 6 Seikei University 7 Shizuoka University 8 Nara Institute of Science and Technology 9 Ritsumeikan University 10(More)
A blind dereverberation method based on power spectral subtraction (SS) using a multi-channel least mean squares algorithm was previously proposed to suppress the reverberant speech without additive noise. The results of isolated word speech recognition experiments showed that this method achieved significant improvements over conventional cepstral mean(More)
In this study, we investigate the e ectiveness of an unknown word processing(UWP) algorithm, which is incorporated into an N-gram language model based speech recognition system for dealing with lled pauses and outof-vocabulary(OOV) words. We have already been investigated the e ect of the UWP algorithm, which utilizes a simple subword sequence decoder, in a(More)
In this paper we propose bottleneck features of deep neural network for distant-talking speaker identification. The accuracy of distant-talking speaker recognition is significantly degraded under reverberant environment. Feature mapping or feature transformation has been shown efficacy in channel-mismatch speaker recognition. Bottleneck feature derived from(More)
An architecture for highly-interactive human-like spoken-dialog agent is discussed in this paper. In order to easily integrate the modules of different characteristics including speech recognizer, speech synthesizer, facial-image synthesizer and dialog controller, each module is modeled as a virtual machine that has a simple common interface and is(More)