Yoshinobu Kajikawa

Yoshinobu Kajikawa

Kansai University

H-index: 21

Asia-Japan

About Yoshinobu Kajikawa

Yoshinobu Kajikawa, With an exceptional h-index of 21 and a recent h-index of 14 (since 2020), a distinguished researcher at Kansai University, specializes in the field of Acoustic Signal Processing, Audio and Acoustic Transducer, Active Noise Control, Spatial Audio, Linearization of Loudspeakers.

His recent articles reflect a diverse array of research interests and contributions to the field:

騒音制御フィルタ選択によるバーチャルセンシングフィードバック ANC システムに関する検討

ANC システムにおけるバーチャルセンシング技術の消音領域に関する検討

Sound Source Localization for a Source inside a Structure using Ac-CycleGAN

能動騒音制御分野への入門のための文献 6 選

スライディング DCT 入力 2D-CNN を用いた固定フィルタ選択 ANC システム

Low-Complexity and Accurate Noise Suppression Based on an a Priori SNR Model for Robust Speech Recognition on Embedded Systems and Its Evaluation in a Car Environment

Hybrid Active Noise Control with Auxiliary Filter-based Virtual Sensing

Editorial for the Special Issue on Advanced Acoustic, Sound and Audio Processing Techniques and Their Applications

Yoshinobu Kajikawa Information

University

Kansai University

Position

Professor of Electrical and Electronic Engineering

Citations(all)

2219

Citations(since 2020)

981

Cited By

1548

hIndex(all)

21

hIndex(since 2020)

14

i10Index(all)

59

i10Index(since 2020)

20

Email

University Profile Page

Kansai University

Yoshinobu Kajikawa Skills & Research Interests

Acoustic Signal Processing

Audio and Acoustic Transducer

Active Noise Control

Spatial Audio

Linearization of Loudspeakers

Top articles of Yoshinobu Kajikawa

騒音制御フィルタ選択によるバーチャルセンシングフィードバック ANC システムに関する検討

Authors

豊岡祥太, 梶川嘉延

Journal

研究報告音声言語情報処理 (SLP)

Published Date

2024/2/22

論文抄録本稿では, 騒音制御フィルタ選択によるバーチャルセンシングフィードバック ANC システムを提案する. 提案 ANC システムでは, 騒音に対して最適な騒音制御フィルタを選択することによって, 騒音変化に素早く追従することが可能である. システム内部で得られる参照信号を帯域の異なる複数のバンドパスフィルタに通し, 騒音のパワーが強い帯域に対して最適な騒音制御フィルタを選択することで, 騒音変化に追従可能である. 実際のインパルス応答を用いたシミュレーションにより, 騒音が変化しても約 18dB の騒音低減が可能であることを示す.

ANC システムにおけるバーチャルセンシング技術の消音領域に関する検討

Authors

豊岡祥太, 梶川嘉延

Journal

IEICE Conferences Archives

Published Date

2023/9/5

アクティブノイズコントロール(ANC)システムは騒音に対して同振幅,逆位相の擬似騒音を生成し,騒音と重ね合わせることで騒音低減を行うシステムである.ANC では一般的に誤差マイクロホン地点を中心に消音領域(ZoQ : Zone of Quiet) を生成できるが,ZoQ の大きさは制御対象となる騒音の周波数で決まる.よって,所望地点にマイクロホンを設置できない場合,十分な低減効果が得られない可能性がある.このような状況に対応するためには,バーチャルセンシング技術を導入しZoQ を所望地点に移動させる必要がある.本稿では,バーチャルセンシングを導入したANCシステムについて実機実験を行い,バーチャルセンシングを用いた場合の消音領域の大きさについて報告する.

Sound Source Localization for a Source inside a Structure using Ac-CycleGAN

Authors

Shunsuke Kita,Choong Sik Park,Yoshinobu Kajikawa

Journal

arXiv preprint arXiv:2312.04846

Published Date

2023/12/8

We propose a method for sound source localization (SSL) for a source inside a structure using Ac-CycleGAN under unpaired data conditions. The proposed method utilizes a large amount of simulated data and a small amount of actual experimental data to locate a sound source inside a structure in a real environment. An Ac-CycleGAN generator contributes to the transformation of simulated data into real data, or vice versa, using unpaired data from both domains. The discriminator of an Ac-CycleGAN model is designed to differentiate between the transformed data generated by the generator and real data, while also predicting the location of the sound source. Vectors representing the frequency spectrum of the accelerometers (FSAs) measured at three points outside the structure are used as input data and the source areas inside the structure are used as labels. The input data vectors are concatenated vertically to form an image. Labels are defined by dividing the interior of the structure into eight areas with one-hot encoding for each area. Thus, the SSL problem is redefined as an image-classification problem to stochastically estimate the location of the sound source. We show that it is possible to estimate the sound source location using the Ac-CycleGAN discriminator for unpaired data across domains. Furthermore, we analyze the discriminative factors for distinguishing the data. The proposed model exhibited an accuracy exceeding 90\% when trained on 80\% of actual data (12.5\% of simulated data). Despite potential imperfections in the domain transformation process carried out by the Ac-CycleGAN generator, the discriminator can …

能動騒音制御分野への入門のための文献 6 選

Authors

梶川嘉延, 雉本信哉, 穴井謙

Journal

騒音制御= The journal of the INCE of Japan

Published Date

2023

トクシュウ ソウオン・シンドウ ケンキュウシャ ガ ソウオン・シンドウ ケンキュウシャ ニ ススメル ブンケン 128 セン

スライディング DCT 入力 2D-CNN を用いた固定フィルタ選択 ANC システム

Authors

土井健矢, 梶川嘉延

Journal

研究報告音声言語情報処理 (SLP)

Published Date

2023/2/21

論文抄録本稿ではスライディング離散コサイン変換 (Sliding DCT) 入力 2 次元畳み込みニューラルネットワーク (2D-CNN) による固定フィルタ選択アクティブノイズコントロール (SFANC) システムを提案する. 提案法では, 騒音信号の CNN への入力前処理にスライディング DCT を用いることで, SFANC システムにおける騒音制御フィルタの切り替えを高速化する. スライディング DCT は再帰的に周波数領域表現への変換が可能であるため, これまで音響クラス分類の信号前処理に用いられてきた短時間フーリエ変換 (STFT) によるスペクトログラム入力よりも演算を高速化できると考えられる. 同様の手法でスライディング DCT の他にもスライディング離散フーリエ変換 (Sliding DFT) があるが, こちらは DCT と異なり実数値を複素数値に変換するため, CNN への入力時に絶対値を取る演算が追加で必要となる. スライディング DCT とスペクトログラムそれぞれを用いた場合における CNN の分類精度は同程度であり, CNN への入力前処理に要する時間は, スライディング DCT 入力では約 0.0058 ms で, スペクトログラム入力の約 1/240 である.

Low-Complexity and Accurate Noise Suppression Based on an a Priori SNR Model for Robust Speech Recognition on Embedded Systems and Its Evaluation in a Car Environment

Authors

Masanori Tsujikawa,Yoshinobu Kajikawa

Journal

IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

Published Date

2023/9/1

In this paper, we propose a low-complexity and accurate noise suppression based on an a priori SNR (Speech to Noise Ratio) model for greater robustness w.r.t. short-term noise-fluctuation. The a priori SNR, the ratio of speech spectra and noise spectra in the spectral domain, represents the difference between speech features and noise features in the feature domain, including the mel-cepstral domain and the logarithmic power spectral domain. This is because logarithmic operations are used for domain conversions. Therefore, an a priori SNR model can easily be expressed in terms of the difference between the speech model and the noise model, which are modeled by the Gaussian mixture models, and it can be generated with low computational cost. By using a priori SNRs accurately estimated on the basis of an a priori SNR model, it is possible to calculate accurate coefficients of noise suppression filters …

Hybrid Active Noise Control with Auxiliary Filter-based Virtual Sensing

Authors

Shota TOYOOKA,Yoshinobu KAJIKAWA

Journal

INTER-NOISE and NOISE-CON Congress and Conference Proceedings

Published Date

2023/11/30

In this paper, we propose a hybrid active noise control system with auxiliary filter-based virtual sensing. The proposed system consists of a Feed-Forward (FF) part, which can reduce broadband noise, and a FeedBack (FB) part, which can reduce narrowband disturbance components, by introducing virtual sensing into the hybrid ANC. Virtual sensing is realized by using two auxiliary filters because the optimal noise control filters are different for the FF and FB parts, respectively. These auxiliary filters store information on the optimal noise control filters that can reduce both narrowband disturbance and broadband noise at the desired location in the tuning stage. Through simulations with real impulse responses for real applications, we demonstrate that the proposed hybrid ANC can achieve an overall noise reduction of about 30 dB.

Editorial for the Special Issue on Advanced Acoustic, Sound and Audio Processing Techniques and Their Applications

Authors

Yu Tsao,Shoji Makino,Yoshinobu Kajikawa,Nobutaka Ono

Journal

APSIPA Transactions on Signal and Information Processing

Published Date

2023

With recent advancements in sensing, computing, and communication capabilities, substantial quantities of acoustic, sound, and audio (ASA) data have become conveniently accessible. The extensive and diverse dataset presents an opportunity to develop systems for a wide range of applications using state-of-the-art artificial intelligence (AI) algorithms. Despite the proliferation of AI-driven systems utilizing ASA data and system frameworks, there is still untapped potential for improving performance and exploring novel directions. This special issue focuses on all aspects of pattern recognition, information retrieval, and front-end processing (enhancement, separation, and noise cancellation) of ASA signals. This special issue has collected nine excellent articles reviewed and highly recommended by the editors and reviewers. The first paper is “Movable Virtual Sound Source Construction Based on Wave Field Synthesis using a Linear Parametric Loudspeaker Array,” authored by Yuting Geng, Shiori Sayama, Masato Nakayama, and Takanobu Nishiura. This paper outlines a novel approach for constructing a movable virtual sound source (VSS) using a linear arrangement of parametric loudspeakers. Experimental findings indicate that, in comparison to a VSS generated using conventional electro-dynamic loudspeakers, the proposed method can

Active noise control system utilizing noise cancellation sounds

Published Date

2023/4/4

Adaptive operations of a first noise control system and a second noise control system may include a speaker that outputs noise cancellation sound, a microphone that detects an error signal, an auxiliary filter that generates, from a noise signal, a correction signal that corrects the error signal so that a difference in a position between the microphone and a noise cancellation position is compensated, and an adaptive filter that performs an adaptive operation using the corrected error signal to generate the noise cancellation sound from the noise signal are alternately performed. A transfer function learned in a state in which the second noise control system is stopped is set in the auxiliary filter of the first noise control system, and a transfer function learned in a state in which the adaptive operation of the first noise control system is stopped is set in the auxiliary filter of the second noise control system.

Study on sound source localization inside a structure using a domain transfer model for real-world adaption of a trained model

Authors

Shunsuke Kita,Yoshinobu Kajikawa

Journal

INTER-NOISE and NOISE-CON Congress and Conference Proceedings

Published Date

2023/2/1

Sound source localization (SSL) is important to reduce noise of products such as machinery and electrical appliances. Currently, SSL methods have been proposed that use the correlation of time-frequency signals of sound waves observed by multiple microphones. However, the application of these SSL methods are limited to the case where the acoustic signals can be directly observed. In the case of estimating the location of a sound source inside a structure from outside the structure, these methods are not applicable because the acoustic signal is observed as indirect sound. An SSL method using deep neural network (DNN) and computer-aided engineering (CAE) was proposed to estimate sound sources inside structures. This method successfully estimates the location of sound sources inside a structure from signals observed by accelerometers installed on the outer surface of the structure in both CAE and …

Performance Improvement of Remote Microphone-Based Virtual Sensing Method for Feedforward Active Noise Control System with Coherence-Adjusting Filter

Authors

Tianyu Xie,Shota Toyooka,Kenta Iwai,Yoshinobu Kajikawa

Journal

IEICE Proceedings Series

Published Date

2023/8/31

In this paper, we propose a new Remote Microphone-based Virtual Sensing (RMVS) method for feedforward active noise control (ANC) system with two coherence adjusting filters (CAF). The feedforward ANC system with RMVS has a poor noise reduction performance under low coherence conditions between signals obtained at reference microphone, error microphone and virtual microphone. The CAF can improve the coherences between the signals obtained at three microphones and prevents the degradation of noise reduction. Unlike the basic feedforward ANC system, the feedforward ANC system with RMVS has three microphone; reference microphone, error microphone and virtual microphone. The low coherence in the combinations of the three microphones yields the degradation of noise reduction performance of the feedforward ANC system with RMVS. Hence, the proposed system utilizes two CAFs in …

Active noise control system

Published Date

2023/11/9

An active noise control system (500) includes a structure (80) and a plurality of piezoelectric speakers (10). The piezoelectric speakers (10) are disposed on a surface (80 s) of the structure (80). The piezoelectric speakers (10) each have a radiation surface extending along a first direction (D1) and a second direction (D2). The first direction (D1) is a direction along which centers of the radiation surfaces of the piezoelectric speakers (10) are arranged so that the piezoelectric speakers (10) are adjacent to each other. The second direction (D2) is a direction orthogonal to the first direction (D1). The radiation surface of each of the piezoelectric speakers (10) is shorter in a dimension (L1) in the first direction (D1) than in a dimension (L2) in the second direction (D2).

A Study on Single Modal Ear Acoustic Personal Authentication System

Authors

Sora Takagi,Shunsuke Kita,Yoshinobu Kajikawa

Journal

IEICE Technical Report; IEICE Tech. Rep.

Published Date

2023/2/22

(in English) Biometric authentication requires the provision of the user's personal biometric information. Currently, the mainstream biometric authentication is fingerprint recognition and face recognition, but these have a high privacy of biometric information used, and there is resistance to providing biometric information. In this paper, we examine a personal authentication system that uses acoustic signal related to pinna. We believe that using these methods will make it possible to use biometric authentication with less privacy of the information provided, because authentication can be performed using invisible information. We have studied an ear acoustic personal authentication system using the Pinna Related Transfer Function (PRTF) measured from the pinna, but if the position of the measuring device fluctuates during the measurement, the accuracy of the authentication will be reduced. In this paper, we propose …

2 次元空間の騒音源移動を考慮した ANC システムに関する検討

Authors

豊岡祥太, 梶川嘉延

Journal

聴覚研究会資料= Proceedings of the auditory research meeting

Published Date

2023

2次元空間の騒音源移動を考慮したANCシステムに関する検討 | CiNii Research CiNii 国立情報学 研究所 学術情報ナビゲータ[サイニィ] 詳細へ移動 検索フォームへ移動 論文・データをさがす 大学 図書館の本をさがす 日本の博士論文をさがす English 検索 タイトル 人物/団体名 所属機関 ISSN DOI 期間 ~ 本文リンク 本文リンクあり データソース JaLC IRDB Crossref DataCite NDL NDL-Digital RUDA JDCat NINJAL CiNii Articles CiNii Books CiNii Dissertations DBpedia Nikkei BP KAKEN Integbio MDR PubMed LSDB Archive 公共データカタログ ムーンショット型研究開発事業 すべて 研究データ 論文 本 博士論文 プロジェクト 2次元空間の騒音源移動を考慮したANCシステム に関する検討 豊岡祥太 梶川 嘉延 書誌事項 タイトル別名 2ジゲン クウカン ノ ソウオン ゲン イドウ オ コウリョ シタ ANC システム ニ カンスル ケントウ Study on ANC System Considering Noise Source Movement in 2-D Space この論文をさがす NDL ONLINE CiNii Books 収録刊行物 聴覚研究会…

前後方向マイクロホン配置を用いた車載用高性能マイクロホンアレーシステム

Authors

辻川剛範, 杉山昭彦, 花沢健, 梶川嘉延

Journal

電子情報通信学会論文誌 A

Published Date

2023/2/1

前後方向マイクロホン配置を用いた車載用高性能マイクロホンアレーシステムを提案する.車室内でマイクロホン配置に課せられる制約を利用して,業界標準の左右方向マイクロホンアレー軸を90度回転して前後方向とすることにより,雑音源方向に投影されたアレー軸に関する話者方向指向性の虚像を,雑音源と異なる方向に向ける.話者方向指向性の虚像が雑音方向と重なることがないので,音声と雑音を方向で区別することが可能となり,少ない音声歪で高い雑音抑圧度を達成できる.最初に,左右方向のアレー軸を有する2マイクロホンアレーで,アレー軸に関して投影された話者方向指向性の虚像に重なる方向の車室内雑音が抑圧できないことを,マイクロホン信号間相対位相の解析を通じて示す.続いて,車室内で録音した信号によるシミュレーション結果を用いて,提案するマイクロホンアレーシステムが従来のマイクロホンアレーシステムより雑音抑圧量を最大7dB,単語正解精度を最大13%改善することを示す.

アクティブノイズコントロールにおける最近の動向

Authors

梶川嘉延

Journal

電子情報通信学会 基礎・境界ソサイエティ Fundamentals Review

Published Date

2023/7/1

本稿では不快な騒音を音で制御・低減するアクティブノイズコントロール (ANC) について, これまでの技術の変遷を展望するとともに最近の動向について解説する. ANC はスピーカ (二次音源) からの制御音により騒音源 (一次音源) からの騒音を制御・低減する技術であり, 近年ではノイズキャンセリングヘッドホンなどを通じて一般にもその技術が広く知られるようになった. しかしながら, オフィスやイベント会場などの公共空間や車内や住宅内などの交通・住環境において, 三次元的に騒音を制御することは未だ広くは実用化されていない. このような音空間において三次元的に騒音制御を実現するためには乗り越えるべき課題が多々残されている. 本稿では, まず ANC システムを実装する上で重要となる消音領域 (ZoQ) を任意の位置に移動させるバーチャルセンシングについて, 代表的な 3 手法を紹介するとともに, それらの性能比較を示す. また, フィードフォワード ANC システムにおいて重要となる因果性制約について説明し, その 1 解決策であるオーバーサンプリングを利用した ANC システムについて述べる. そして, 近年様々な分野において利用されている機械学習の ANC システムへの適用方法について紹介する.

Head-mounted multi-channel feedforward active noise control system for reducing noise arriving from various directions

Authors

Takumi Miyake,Kenta Iwai,Yoshinobu Kajikawa

Journal

IEEE Access

Published Date

2023/1/18

A head-mounted active noise control (ANC) system based on the single-channel ( ) 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 ( ) 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 …

ハイブリッド ANC システムにおけるバーチャルセンシング法の検討

Authors

豊岡祥太, 梶川嘉延

Journal

研究報告音声言語情報処理 (SLP)

Published Date

2023/2/21

論文抄録本稿では, バーチャルセンシングの一つである補助フィルタ (AFVS) 法を用いたハイブリッド ANC システムを提案する. この提案法では, 広帯域騒音を低減可能なフィードフォワード (FF) 部と, 狭帯域の外乱成分を低減可能なフィードバック (FB) 部から構成されているハイブリッド ANC システムに, バーチャルセンシングを導入する. FF 部と FB 部で最適な騒音制御フィルタは異なるため, 従来の FF, FB 構成の Tuning Stage にて 2 つの補助フィルタを作成し, バーチャルセンシングを実現する. 実際のインパルス応答を用いたシミュレーションにより, 提案法では全体で約 30dB の騒音低減が可能なことを示す.

Linear Microphone Array Parallel to the Driving Direction for in-Car Speech Enhancement

Authors

Masanori Tsujikawa,Akihiko Sugiyama,Ken Hanazawa,Yoshinobu Kajikawa

Published Date

2023/6/4

This paper proposes a linear microphone array parallel to the driving direction for in-car speech enhancement. In contrast to other linear microphone arrays in the car cabin reported in a literature or implemented as a commercial product, the array axis is arranged in parallel to the driving direction. Thanks to the 90°-rotated array axis with the constraints on the microphone position specific to the car environment, a mirror image of the directivity toward the talker with respect to the array axis is no longer projected in the direction of interference and redirected to a direction with no interference. As a result, the talker speech can be discriminated from the interference by directivity, leading to good interference reduction with little speech distortion. Simulation results with signals recorded in a car environment show that the proposed linear microphone array with the array axis parallel to the driving direction has a null in the …

SFANC with Compensation Filter Based on MEFxDCTLMS Algorithm

Authors

Kenya Doi,Yoshinobu Kajikawa

Published Date

2023/10/31

This paper proposes a hybrid system of the selective fixed-filters active noise control (SFANC) and the modified error filtered-x DCT-LMS (MEFxDCTLMS) algorithm. Conventional SFANC systems have a problem that the noise reduction performance deteriorates due to the incorrect selection of fixed filters. In addition, the conventional SFANC system uses the spectrogram by short-time Fourier transform (STFT) of the reference signal detected by the reference microphone as an input in a two-dimensional convolutional neural network (2D-CNN). However, the processing delay for calculating the spectrogram delays the fixed filter selection in SFANC, resulting in degraded noise reduction performance. To compensate for the fixed filter, we propose a system in which an adaptive filter based on MEFxDCTLMS is vertically connected to SFANC. This improves the degradation of the noise reduction performance due to the …

See List of Professors in Yoshinobu Kajikawa University(Kansai University)

Yoshinobu Kajikawa FAQs

What is Yoshinobu Kajikawa's h-index at Kansai University?

The h-index of Yoshinobu Kajikawa has been 14 since 2020 and 21 in total.

What are Yoshinobu Kajikawa's top articles?

The articles with the titles of

騒音制御フィルタ選択によるバーチャルセンシングフィードバック ANC システムに関する検討

ANC システムにおけるバーチャルセンシング技術の消音領域に関する検討

Sound Source Localization for a Source inside a Structure using Ac-CycleGAN

能動騒音制御分野への入門のための文献 6 選

スライディング DCT 入力 2D-CNN を用いた固定フィルタ選択 ANC システム

Low-Complexity and Accurate Noise Suppression Based on an a Priori SNR Model for Robust Speech Recognition on Embedded Systems and Its Evaluation in a Car Environment

Hybrid Active Noise Control with Auxiliary Filter-based Virtual Sensing

Editorial for the Special Issue on Advanced Acoustic, Sound and Audio Processing Techniques and Their Applications

...

are the top articles of Yoshinobu Kajikawa at Kansai University.

What are Yoshinobu Kajikawa's research interests?

The research interests of Yoshinobu Kajikawa are: Acoustic Signal Processing, Audio and Acoustic Transducer, Active Noise Control, Spatial Audio, Linearization of Loudspeakers

What is Yoshinobu Kajikawa's total number of citations?

Yoshinobu Kajikawa has 2,219 citations in total.

What are the co-authors of Yoshinobu Kajikawa?

The co-authors of Yoshinobu Kajikawa are Bhan Lam, CY Chang, Dongyuan Shi, Chuang Shi, Seiji Miyoshi, Kenta Iwai.

    Co-Authors

    H-index: 23
    Bhan Lam

    Bhan Lam

    Nanyang Technological University

    H-index: 22
    CY Chang

    CY Chang

    Chung Yuan Christian University

    H-index: 21
    Dongyuan Shi

    Dongyuan Shi

    Nanyang Technological University

    H-index: 21
    Chuang Shi

    Chuang Shi

    University of Electronic Science and Technology of China

    H-index: 11
    Seiji Miyoshi

    Seiji Miyoshi

    Kansai University

    H-index: 5
    Kenta Iwai

    Kenta Iwai

    Ritsumeikan University

    academic-engine

    Useful Links