Kenta Iwai

Dr. Kenta Iwai


Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 January 2023 in IEEE Access
Verified
Journal Article
Open access
TMTakumi Miyake
Takumi Miyake
KIKenta Iwai
Kenta Iwai
Yoshinobu KajikawaYoshinobu Kajikawa
Yoshinobu Kajikawa
A head-mounted active noise control (ANC) system based on the single-channel (1x1x1) feedforward control has been studied as a noise reduction method in complicated noise environments such as factories in which various machines are located that generate unwanted noises. However, it is difficult to reduce the noise arriving from various directions because the causality constraint cannot be satisfied. To solve this problem, we developed a head-mounted ANC system based on the multi-channel (2x1x1) feedforward control. This ANC system has two reference microphones that satisfy the causality constraint. Therefore, the proposed head-mounted ANC system achieves the expansion of noise reducible range of 0-100° compared to that of the conventional head-mounted ANC with the range of 0-60°. In this paper, we examine the effectiveness of the newly developed head-mounted ANC system. In the experiments, we evaluate the effectiveness of placing two reference microphones on noise reduction performance in the case of various arrival direction of the unwanted noise. Experimental results demonstrate that the proposed system increases the noise reducible range and improves the noise reduction performance for two noise sources. In addition, we theoretically prove the effectiveness of the proposed system.

20 April 2023 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Verified
Journal Article
Open access
NTNguyen Binh Thien
Nguyen Binh Thien
YWYukoh Wakabayashi
Yukoh Wakabayashi
KIKenta Iwai
Kenta Iwai
TNTakanobu Nishiura
Takanobu Nishiura
This paper presents improvements to two-stage algorithms for estimating the short-time Fourier transform (STFT) phase from only the amplitude by using deep neural networks (DNNs). The phase is difficult to reconstruct due to its sensitivity to the waveform shift and wrapping issue. To mitigate these problems, two-stage approaches indirectly estimate the phase through phase derivatives, i.e., instantaneous frequency (IF) and group delay (GD). In the first stage, the IF and GD are estimated from the amplitude using DNNs, and then in the second stage, the phase is reconstructed by maintaining the IF/GD information. Conventional methods for the second stage do not consider the importance of high-amplitude time–frequency bins, e.g., the least squares-based method, or lack a solid model, e.g., the average-based method. To address these problems, we propose improvements to the second stage of two-stage algorithms by using von Mises distribution-based maximum likelihood and weighted least squares. We also provide theoretical discussions for the phase reconstruction, including the investigations of the properties of the GD and roles of the IF/GD information in the inverse STFT. On the basis of the analysis, we propose a new phase-based feature, i.e., inter-frequency phase difference (IFPD), and demonstrate its application in two-stage phase reconstruction algorithms. We conducted subjective and objective experiments to compare the performances of our proposed and conventional methods. The results confirm that the proposed method using the IFPD performs better than other methods for all metrics.

19 November 2018 in Applied Sciences
Verified
Journal Article
Open access
PDF file
KIKenta Iwai
Kenta Iwai
SHSatoru Hase
Satoru Hase
Yoshinobu KajikawaYoshinobu Kajikawa
Yoshinobu Kajikawa
In this paper, we propose a multichannel active noise control (ANC) system with an optimal reference microphone selector based on the time difference of arrival (TDOA). A multichannel feedforward ANC system using upstream reference signals can reduce various noises such as broadband noise by arranging reference microphones close to noise sources. However, the noise reduction performance of an ANC system degrades when the noise environment changes, such as the arrival direction. This is because some reference microphones do not satisfy the causality constraint that the unwanted noise propagates to the control point faster than the anti-noise used to cancel the unwanted noise. To solve this problem, we propose a multichannel ANC system with an optimal reference microphone selector. This selector chooses the reference microphones that satisfy the causality constraint based on the TDOA. Some experimental results demonstrate that the proposed system can choose the optimal reference microphones and effectively reduce unwanted acoustic noise.

4 October 2024 in Information
Verified
Journal Article
Open access
PDF file
PCPeng CHEN
Peng CHEN
BNBinh Thien Nguyen
Binh Thien Nguyen
KIKenta Iwai
Kenta Iwai
TNTakanobu Nishiura
Takanobu Nishiura
An effective approach to addressing the speech separation problem is utilizing a time–frequency (T-F) mask. The ideal binary mask (IBM) and ideal ratio mask (IRM) have long been widely used to separate speech signals. However, the IBM is better at improving speech intelligibility, while the IRM is better at improving speech quality. To leverage their respective strengths and overcome weaknesses, we propose an ideal threshold-based mask (ITM) to combine these two masks. By adjusting two thresholds, these two masks are combined to jointly act on speech separation. We list the impact of using different threshold combinations on speech separation performance under ideal conditions and discuss a reasonable range for fine tuning the thresholds. By using masks as a training target, to evaluate the effectiveness of the proposed method, we conducted supervised speech separation experiments applying a deep neural network (DNN) and long short-term memory (LSTM), the results of which were measured by three objective indicators: the signal-to-distortion ratio (SDR), signal-to-interference ratio (SIR), and signal-to-artifact ratio improvement (SAR). Experimental results show that the proposed mask combines the strengths of the IBM and IRM and implies that the accuracy of speech separation can potentially be further improved by effectively leveraging the advantages of different masks.

2 February 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
CCChengkai Cai
Chengkai Cai
KIKenta Iwai
Kenta Iwai
TNTakanobu Nishiura
Takanobu Nishiura
The development of distant-talk measurement systems has been attracting attention since they can be applied to many situations such as security and disaster relief. One such system that uses a device called a laser Doppler vibrometer (LDV) to acquire sound by measuring an object’s vibration caused by the sound source has been proposed. Different from traditional microphones, an LDV can pick up the target sound from a distance even in a noisy environment. However, the acquired sounds are greatly distorted due to the object’s shape and frequency response. Due to the particularity of the degradation of observed speech, conventional methods cannot be effectively applied to LDVs. We propose two speech enhancement methods that are based on two-stage processing with deep neural networks for LDVs. With the first proposed method, the amplitude spectrum of the observed speech is first restored. The phase difference between the observed and clean speech is then estimated using the restored amplitude spectrum. With the other proposed method, the low-frequency components of the observed speech are first restored. The high-frequency components are then estimated by the restored low-frequency components. The evaluation results indicate that they improved the observed speech in sound quality, deterioration degree, and intelligibility.

6 December 2023 in APSIPA Transactions on Signal and Information Processing
Verified
Journal Article
Open access
PDF file
KIKenta Iwai
Kenta Iwai
TNTakanobu Nishiura
Takanobu Nishiura
In this paper, a feedforward active noise control (ANC) system with an optical laser microphone is proposed and the effectiveness of the proposed system is introduced through real-world experiments. The main problem with the feedforward ANC system is the degradation of its noise reduction performance due to the causality constraint, that is, the ANC system should update the noise control filter and emit the anti-noise until unwanted noise reaches the error microphone. As a solution, the proposed ANC system utilizes the optical laser microphone as the reference microphone. The optical laser microphone picks up the vibration of the noise source by emitting an optical laser beam to the surface of the vibrating object. Since the reference signal obtained by the optical laser microphone is different from the unwanted noise obtained by the ordinary microphone as the error sensor, the proposed system utilizes the first-order differentiator to convert the vibration velocity into the acceleration for the improvement of the coherence between the unwanted noises obtained by the reference and error microphones. Simulation and experimental results show that the proposed system can reduce the unwanted noise by almost the same amount as the conventional system while the causality constraint is relaxed.