Automated testing of basic recognition capability for speech recognition systemsFutoshi IwamaTakashi Fukuda2019ICST 2019
Data augmentation improves recognition of foreign accented speechTakashi FukudaRaul Fernandezet al.2018INTERSPEECH 2018
Efficient knowledge distillation from an ensemble of teachersTakashi FukudaMasayuki Suzukiet al.2017INTERSPEECH 2017
Effective joint training of denoising feature space transforms and Neural Network based acoustic modelsTakashi FukudaOsamu Ichikawaet al.2017ICASSP 2017
Harmonic feature fusion for robust neural network-based acoustic modelingOsamu IchikawaTakashi Fukudaet al.2017ICASSP 2017
Regularized feature-space discriminative adaptation for robust ASRTakashi FukudaOsamu Ichikawaet al.2014INTERSPEECH 2014
Constructing ensembles of dissimilar acoustic models using hidden attributes of training dataTakashi FukudaRyuki Tachibanaet al.2012ICASSP 2012
Model-based noise reduction leveraging frequency-wise confidence metric for in-car speech recognitionOsamu IchikawaSteven J. Rennieet al.2012ICASSP 2012