Mario Kaoru Watanabe

Learn More
Partial phase reconstruction based on a confidence domain has recently been shown to provide improved signal reconstruction performance in a single-channel source separation scenario. In this paper, we replace the previous binarized fixed-threshold confidence domain with a new signal-dependent one estimated by employing a sinusoidal model to be applied on(More)
Many short-time Fourier transform (STFT) based singlechannel speech enhancement algorithms are focused on estimating clean speech spectral amplitude from the noisy observed signal in order to suppress the additive noise. To this end, the state-of-the-art speech enhancement algorithms, employ noisy amplitude information and correspondingly a priori and a(More)
Many short-time Fourier transform (STFT) based singlechannel speech enhancement algorithms are focused on estimating clean speech spectral amplitude from the noisy observed signal in order to suppress the additive noise. To this end, the state-of-the-art speech enhancement algorithms, employ noisy amplitude information and correspondingly a priori and a(More)
While the state-of-the-art speech enhancement methods are focused on the modification of the noisy spectral amplitude, our recent findings demonstrate positive impact of incorporating the speech phase spectrum in speech enhancement. In this show and tell proposal, we demonstrate the recent progress towards utilizing the phase information in closed-loop(More)
In many artificial intelligence systems human voice is considered as the medium for information transmission. Human-machine communication by voice becomes difficult when speech is mixed with some background noise. As a remedy, a single-channel speech enhancement is indispensable for reducing background noise from noisy speech to make it suitable for(More)
While the state-of-the-art speech enhancement methods are focused on the modification of the noisy spectral amplitude, our recent findings demonstrate positive impact of incorporating the speech phase spectrum in speech enhancement. In this show and tell proposal, we demonstrate the recent progress towards utilizing the phase information in closed-loop(More)
  • 1