Sound source segregation based on estimating incident angle of each frequency component of input signals acquired by multiple microphones

スポンサーリンク

概要

• 論文の詳細を見る

• We have developed a method of segregating desired speech from concurrent sounds received by two microphones. In this method, which we call SAFIA, signals received by two microphones are analyzed by discrete Fourier transformation. For each freqency component, differences in the amplitude and phase between channels are calculated. These differences are used to select frequency components of the signal that come from the desired direction and to reconstruct these components as the desired source signal. To clarify the effect of frequency resolution on the proposed method, we conducted three experiments. First, we analyzed the relationship between frequency resolition and the power spectrum's cumulative distribution. We found that the speech-signal power was concentrated on specific frequency components when the frequency resolution was about 10Hz. Second, we determined whether a given frequency resolution decreased the overlap between the frequency components of two speech signals. A 10-Hz frequency resolution minimized the overlap. Third, we analyzed the relationship between sound quality and frequency resolution through subjective tests. The best frequency resolution in terms of sound quality corresponded to the frequency resolutions that concentrated the speech signal power on specific frequency components and that minimized the degree of overlap. Finally, we demonstrated that this method improved the signal-to-noise ratio by over 18 dB.

• 社団法人日本音響学会の論文

著者

• Kaneda Yutaka

  Department Of Information And Communication Engineering Graduate School Of Eng Tokyo Denki Universit

• Kaneda Yutaka

  Department Of Information And Communication Engineering Tokyo Denki University

• Aoki Shigeaki

  Media Technology Development Center Ntt Communications Corporation

• Aoki Mariko

  Media Processing Project, NTT Cyber Space Laboratories

• Okamoto Manabu

  Business Communications Headquarters, NTT East Corporation

• Matsui Hiroyuki

  Solution Business Division, NTT Communications Corporation

• Sakurai Tetsuma

  Department of Information Science, Faculty of Engineering, Fukui University

• Aoki Mariko

  Media Processing Project Ntt Cyber Space Laboratories

• Okamoto Manabu

  Business Communications Headquarters Ntt East Corporation

• Sakurai T

  Department Of Information Science Faculty Of Engineering Fukui University

• Kaneda Yutaka

  Department Of Chemistry Faculty Of Engineering Science Osaka University

• Matsui H

  Solution Business Division Ntt Communications Corporation

関連論文

• Sound source segregation based on estimating incident angle of each frequency component of input signals acquired by multiple microphones
• Large Third-Order Nonlinear Optical Susceptibilities of Multiply-Bonded M_2(pyphos)_4 and M_2Pd_2Cl_2(pyphos)_4(M=Cr, Mo ; Pyphos=6-Diphenylphosphino-2-pyridonate) by Picosecond Degenerate Four-Wave Mixing Method
• A Flexible and Low-Cost ASIC Line Management Technology Taking Operator's Skill-Level as a Scheduling-Factor into Consideration
• A Scalable and Flexible CIM System with Precise and Quick Scheduler for ASIC
• Study of harmonic distortion on impulse response measurement with logarithmic time stretched pulse
• Improving the robustness of multiple signal classification (MUSIC) method to reflected sounds by sub-band peak-hold processing
• Impulse response measurement that maximizes signal-to-noise ratio against ambient noise

スポンサーリンク