Stereo-input speech recognition using sparseness-based time-frequency masking in a reverberant environment

Yosuke Izumi, Kenta Nishiki, Shinji Watanabe, Takuya Nishimoto, Nobutaka Ono, Shigeki Sagayama