The Experts below are selected from a list of 5097 Experts worldwide ranked by ideXlab platform

Yu Takahashi - One of the best experts on this subject based on the ideXlab platform.

  • superresolution based Stereo Signal separation via supervised nonnegative matrix factorization
    International Conference on Digital Signal Processing, 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • DSP - Superresolution-based Stereo Signal separation via supervised nonnegative matrix factorization
    2013 18th International Conference on Digital Signal Processing (DSP), 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

Hitoshi Kiya - One of the best experts on this subject based on the ideXlab platform.

  • replay attack detection using generalized cross correlation of Stereo Signal
    European Signal Processing Conference, 2019
    Co-Authors: Ryoya Yaguchi, Sayaka Shiota, Nobutaka Ono, Hitoshi Kiya
    Abstract:

    In this paper, we propose a replay attack detection method using the generalized cross-correlation (GCC) of a Stereo Signal for automatic speaker verification. In particular, this method focuses on a specific replay attack characteristics when speech is not active. In a genuine speaker case, when speech is not active, the maximum value of GCC is low since surrounding noise arrives from any direction. In contrast, in a replay attack case, even when the played speech is not active, the maximum value of GCC is high since recorded noise or electromagnetic noise is played by a loudspeaker for replay attack. Based on this assumption, two approaches of replay attack detection are introduced. One is to use the minimum value of GCC in short pauses. The other one is to use the average value of GCC in silent periods before the start point and after the end point of a target utterance. In experiments, it is confirmed that the proposed methods achieve low error rates without environmental restrictions.

  • EUSIPCO - Replay Attack Detection Using Generalized Cross-Correlation of Stereo Signal
    2019 27th European Signal Processing Conference (EUSIPCO), 2019
    Co-Authors: Ryoya Yaguchi, Sayaka Shiota, Nobutaka Ono, Hitoshi Kiya
    Abstract:

    In this paper, we propose a replay attack detection method using the generalized cross-correlation (GCC) of a Stereo Signal for automatic speaker verification. In particular, this method focuses on a specific replay attack characteristics when speech is not active. In a genuine speaker case, when speech is not active, the maximum value of GCC is low since surrounding noise arrives from any direction. In contrast, in a replay attack case, even when the played speech is not active, the maximum value of GCC is high since recorded noise or electromagnetic noise is played by a loudspeaker for replay attack. Based on this assumption, two approaches of replay attack detection are introduced. One is to use the minimum value of GCC in short pauses. The other one is to use the average value of GCC in silent periods before the start point and after the end point of a target utterance. In experiments, it is confirmed that the proposed methods achieve low error rates without environmental restrictions.

Hiroshi Saruwatari - One of the best experts on this subject based on the ideXlab platform.

  • superresolution based Stereo Signal separation via supervised nonnegative matrix factorization
    International Conference on Digital Signal Processing, 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • DSP - Superresolution-based Stereo Signal separation via supervised nonnegative matrix factorization
    2013 18th International Conference on Digital Signal Processing (DSP), 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • blind noise suppression for non audible murmur recognition with Stereo Signal processing
    IEEE Automatic Speech Recognition and Understanding Workshop, 2011
    Co-Authors: Shunta Ishii, Hiroshi Saruwatari, Tomoki Toda, Sakriani Sakti, Satoshi Nakamura
    Abstract:

    In this paper, we propose a blind noise suppression method for Non-Audible Murmur (NAM) recognition. NAM is a very soft whispered voice detected with NAM microphone, which is one of the body-conductive microphones. Due to its recording mechanism, the detected Signal suffers from noise caused by speaker's movements. In the proposed method using a Stereo Signal detected with two NAM microphones, the noise is estimated with blind source separation, and then, spectral subtraction is performed in each channel to reduce the noise. Moreover, channel selection is performed frame by frame to generate less distorted monaural NAM Signal. Experimental results show that 1) word accuracy in large vocabulary continuous NAM recognition is degraded from 69.2% to 53.6% by the noise and 2) it is significantly recovered to 63.3% in a simulated situation and 58.6% in a real situation with the proposed method.

  • ASRU - Blind noise suppression for Non-Audible Murmur recognition with Stereo Signal processing
    2011 IEEE Workshop on Automatic Speech Recognition & Understanding, 2011
    Co-Authors: Shunta Ishii, Hiroshi Saruwatari, Tomoki Toda, Sakriani Sakti, Satoshi Nakamura
    Abstract:

    In this paper, we propose a blind noise suppression method for Non-Audible Murmur (NAM) recognition. NAM is a very soft whispered voice detected with NAM microphone, which is one of the body-conductive microphones. Due to its recording mechanism, the detected Signal suffers from noise caused by speaker's movements. In the proposed method using a Stereo Signal detected with two NAM microphones, the noise is estimated with blind source separation, and then, spectral subtraction is performed in each channel to reduce the noise. Moreover, channel selection is performed frame by frame to generate less distorted monaural NAM Signal. Experimental results show that 1) word accuracy in large vocabulary continuous NAM recognition is degraded from 69.2% to 53.6% by the noise and 2) it is significantly recovered to 63.3% in a simulated situation and 58.6% in a real situation with the proposed method.

  • Compressive coding of Stereo audio Signals extracting sparseness among sound sources with independent component analysis
    IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2007
    Co-Authors: Shigeki Miyabe, Tadashi Mihashi, Hiroshi Saruwatari, Tomoya Takatani, Kiyohiro Shikano, Toshiyuki Nomura
    Abstract:

    In this paper we propose a new compressive coding method of Stereo audio Signals extracting sparseness among sound sources by using independent component analysis. Some researchers have proposed a compressive coding method of multi-channel audio called binaural cue coding (BCC), and the ISO/MPEG standardization group discusses standard of next generation audio based on BCC. BCC has an underlying model assuming existence of only a single sound source in each subband of the multi-channel audio Signal. Mismatch of this model often occurs and as a result quality of reconstructed multi-channel Signal degrades. To extract the time-frequency grids where only a single source exists, we apply independent component analysis (ICA) to Stereo Signals. Using this analysis, a single dominant source can be chosen efficiently in each of frequency bins. In addition, transfer functions to reconstruct Stereo Signal from the dominant source is also extracted by ICA. Experiments based on both objective and subjective evaluations ascertains efficiency of the proposed method.

Daichi Kitamura - One of the best experts on this subject based on the ideXlab platform.

  • superresolution based Stereo Signal separation via supervised nonnegative matrix factorization
    International Conference on Digital Signal Processing, 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • DSP - Superresolution-based Stereo Signal separation via supervised nonnegative matrix factorization
    2013 18th International Conference on Digital Signal Processing (DSP), 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

Kiyohiro Shikano - One of the best experts on this subject based on the ideXlab platform.

  • superresolution based Stereo Signal separation via supervised nonnegative matrix factorization
    International Conference on Digital Signal Processing, 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • DSP - Superresolution-based Stereo Signal separation via supervised nonnegative matrix factorization
    2013 18th International Conference on Digital Signal Processing (DSP), 2013
    Co-Authors: Daichi Kitamura, Hiroshi Saruwatari, Kiyohiro Shikano, Yusuke Iwao, Kazunobu Kondo, Yu Takahashi
    Abstract:

    In this paper, we address a Stereo Signal separation problem and propose a new method utilizing both directional clustering and superresolution-based supervised nonnegative matrix factorization (NMF) via spectrogram extrapolation using supervised bases. In previous studies, a hybrid method concatenating supervised NMF after directional clustering was proposed as for multichannel Signal separation. However, this hybrid method has a problem that the extracted Signal suffers from considerable spectral distortion because directional clustering yields spectral chasms. To solve this problem, we propose a new supervised NMF algorithm that regards the spectral chasms as unseen observations and reconstructs the target source components via spectrogram extrapolation using supervised bases. Our experimental results show that the proposed method outperforms several conventional methods and that the distortion of the extracted Signal can be mitigated by superresolution efficacy.

  • Compressive coding of Stereo audio Signals extracting sparseness among sound sources with independent component analysis
    IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2007
    Co-Authors: Shigeki Miyabe, Tadashi Mihashi, Hiroshi Saruwatari, Tomoya Takatani, Kiyohiro Shikano, Toshiyuki Nomura
    Abstract:

    In this paper we propose a new compressive coding method of Stereo audio Signals extracting sparseness among sound sources by using independent component analysis. Some researchers have proposed a compressive coding method of multi-channel audio called binaural cue coding (BCC), and the ISO/MPEG standardization group discusses standard of next generation audio based on BCC. BCC has an underlying model assuming existence of only a single sound source in each subband of the multi-channel audio Signal. Mismatch of this model often occurs and as a result quality of reconstructed multi-channel Signal degrades. To extract the time-frequency grids where only a single source exists, we apply independent component analysis (ICA) to Stereo Signals. Using this analysis, a single dominant source can be chosen efficiently in each of frequency bins. In addition, transfer functions to reconstruct Stereo Signal from the dominant source is also extracted by ICA. Experiments based on both objective and subjective evaluations ascertains efficiency of the proposed method.