2003 Fiscal Year Annual Research Report

分散して蓄積された音声データを用いて多様な音響モデルを作成する方法の研究

Research Project

Project/Area Number	15200014
Research Category	Grant-in-Aid for Scientific Research (A)
Research Institution	Nagoya University
Principal Investigator	武田一哉名古屋大学, 情報科学研究科, 教授 (20273295)
Co-Investigator(Kenkyū-buntansha)	河原達也京都大学, 学術情報メディアセンター, 教授 (00234104) 鹿野清宏奈良先端科学技術大学院大学, 情報科学研究科, 教授 (00263426)
Keywords	連続音声認識 / 音響モデル / 分散学習 / HMM
Research Abstract	(分散モデル学習環境の構築)音響モデルの分散学習を実行するための,遠隔分散学習サーバプログラム,データ格納サーバプログラムを作成した。データ格納サーバプログラムには(1)データ検索機能(性別,年齢,SN比などで,利用データを検索指定する機能)(2)データ解析機能(発話毎の音素出現率,SN比自動計算する機能)が付与されており,遠隔分散学習プログラムと通信を行うことで,HMMの基本学習プルグラムであるEMアルゴリズムを実行する。サーバ間はSOAP通信方式を利用することで,様々な通信インフラ上での分散学習が可能となっている。さらに,データの秘匿性を向上させるために,分散学習システムのユーザ登録,認証機能も実現されている。当該システムを名古屋大学、京都大学、奈良先端大学院大学の計算機に設置し、分散学習機能の正常動作を確認した。さらに,車内対話音声(名古屋大学),情報案内端末音声(奈良先端大),バス運行間い合わせ対話(京都大),などの学習用音声を,データ格納サーバに格納し遠隔学習の予備実験を開始した。 (モデル混合アルゴリズムの検討)モデル混合,構築アルゴリズムについては,事前分布を仮定するEMアルゴリズムを,単一ガウス分布から混合ガウス分布に拡張し,その性能をシュミレーション実験により確認した。多数話者(500名)の話者依存HMMを作成し話者間性能の分布を分析した。分析結果から,話者別認識率の分散を最小化するためのデータセット構築法に関する検討を開始した。

Research Products

(99 results)

All Other

All Publications (99 results)

[Publications] T.Nishikawa, H.Saruwatari, K.Shikano: "Blind source separation of acoustic signals based on multistage ICA combining frequency-domain ICA and time-domain ICA"IEICE Trans.Fundamentals. Vol.E86-A, No.4. 846-858 (2003)
[Publications] 南條浩輝, 加藤一臣, 李晃伸, 河原達也: "大規模な日本語話し言葉データベースを用いた講演音声認識"電子情報通信学会論文誌. Vol.J86-D-II No.4. 450-459 (2003)
[Publications] T.Nishikawa, H.Saruwatari, K.Shikano: "Stable learning algorithm for blind separation of temporally correlated acoustic signals combining multistage ICA and Linear Prediction"IEICE Trans.Fundamentals. Vol.E86-A, No.8. 2028-2036 (2003)
[Publications] Panikos Heracleous, Satoshi Nakamura, Kiyohiro Shikano: "Simultaneous Recognition of Distant-Talking Speech of Multiple Talkers Based on the 3-D N-best Search"Journal of VLSI Signal Processing, Kluwer Academic Publishers. Vol.36-2. 105-116 (2004)
[Publications] H.Saruwatari, S.Kurita, K.Takeda, F.Itakura, T.Nishikawa.K.Shikano: "Blind Source Separation Combining Independent Component Analysis and Beamforming"EURASIP Journal on Applied Signal Processing. Vol.2003, No.11. 1135-1146 (2003)
[Publications] S.Yamade, A.Lee, H.Saruwatari, K.Shikano: "Unsupervised Speaker Adaptation Based on HMM Sufficient Statistics in Various Noisy Environments"Proceedings of 8th European Conference on Speech Communication and Technology. 1493-1496 (2003)
[Publications] 舛田剛志, 戸田智基, 川波弘道, 猿渡洋, 鹿野清宏: "韻律的に多重化した音声データベースの設計と発話速度におけるその評価"電子情報通信学会論文誌D-II. No.2. 447-455 (2004)
[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Probability Distribution of Time-Series of Speech Spectral Components"IEICE Trans.Fundamentals. Vol.E87-A, No.3. 584-597 (2004)
[Publications] 永田悠, 立蔵洋介, 猿渡洋, 鹿野清宏: "音場再現システムにおける環境変化に適応的な逆フィルターの逐次的緩和アルゴリズム"電子情報通信学会論文誌A. Vol.J86-A, No.8. 824-834 (2003)
[Publications] 西村竜一, 西原洋平, 鶴身玲典, 李晃伸, 猿渡洋, 鹿野清宏: "実環境研究プラットホームとしての音声情報案内システムの運用"電子情報通信学会論文誌. Vol.J87-D-II No.3. 789-798 (2004)
[Publications] Y.Hinamoto, K.Mino, H.Saruwatari, K.Shikano: "Spoken dialogue system based on sound field control and microphone array"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.V. 505-508 (2003)
[Publications] T.Shiraishi, T.Toda, H.Kawanami, H.Saruwatari, K.Shikano: "Simple Designing Methods of Corpus-Based Visual Speech Synthesis"Proceedings of 8th European Conference on Speech Communication and Technology. 2241-2244 (2003)
[Publications] T.Toda, H.Kawai, M.Tsuzaki, K.Shikano: "SEGMENT SELECTION CONSIDERING LOCAL DEGRADATION OF NATURALNESS IN CONCATENATIVE SPEECH SYNTHESIS"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.I. 696-699 (2003)
[Publications] H.Kawanami, Y.Iwami, T.Toda, H.Saruwatari, K.Shikano: "GMM-based Voice Conversion Applied to Emotional Speech Synthesis"Proceedings of 8th European Conference on Speech Communication and Technology. 2401-2404 (2003)
[Publications] Y.Nakajima, H.Kashino, K.Shikano, N.Campbell: "RECOGNITION INPUT INTERFACE USING STETHOSCOPIC MICROPHONE ATTACHED TO THE SKIN"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.V. 708-711 (2003)
[Publications] Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "Stable Learning Algorithm for Blind Separation of Temporally Correlated Signals Combining Multistage ICA and Linear Prediction"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 327-342 (2003)
[Publications] P.Heracleous, S.Nakamura, K.Shikano: "A Semi-Blind Source Separation Method for Hands-Free Speech Recognition of Multiple Talkers"Proceedings of 8th European Conference on Speech Communication and Technology. 509-512 (2003)
[Publications] Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano, Shoko Araki, Shoji Mak: "Multistage ICA for Blind Source Separation of Real Acoustic Convolutive Mixture"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 523-528 (2003)
[Publications] Hiroshi Saruwatari, Tomoya Takatani, Hiroaki Yamajo, Tsuyoki Nishikawa, Kiyohiro Shikano: "Blind Separation and Deconvolution for Real Convolutive Mixture of Temporally Correlated Acoustic Signals Using SIMO-Model-Based ICA"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 549-554 (2003)
[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "SIMO-Model-Based Independent Component Analysis for High-Fidelity Blind Separation of Acoustic Signals"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 993-998 (2003)
[Publications] Hiroshi Saruwatari, Katsuyuki Sawai, Akinobu Lee, Kiyohiro Shikano, Atsunobu Kaminuma, Masao Sakata: "Speech Enhancement and Recognition in Car Environment using Blind Source Separation and Subband Elimination Processing"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 367-372 (2003)
[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Akinobu Lee, Kiyohiro Shikano: "A Fixed-Point ICA Algorithm for Convoluted Speech Signal Separation"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 579-584 (2003)
[Publications] Hiroshi Saruwatari, Hiroaki Yamajo, Tomoya Takatani, Tsuyoki Nishikawa, Kiyohiro Shikano: "Parallel Structured Independent Component Analysis for SIMO-Model-Based Blind Separation and Deconvolution of Convolutive Speech Mixture"Proceedings of International Joint Conference on Neural Networks. 714-719 (2003)
[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "High-Fidelity Blind Separation for Convolutive Mixture of Acoustic Signals Using SIMO-Model-Based Independent Component Analysis"Proceedings of Seventh International Symposium on Signal Processing and its Applications. (2003)
[Publications] Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "Stable Learning Algorithm for Low-Distortion Blind Separation of Real Speech Mixture Combining Multistage ICA AND Linear Prediction"Proceedings of ISCA tutorial and research workshop on non-linear speech processing. 31-34 (2003)
[Publications] R.Nisimura, Y.Nishihara, R.Tsurumi, A.Lee, H.Saruwatari, K.Shikano: "Takemaru-kun : Speech-Oriented Information System for Real World Research Platform"International Workshop on Language Understanding and Agents for Real World Interaction. 70-78 (2003)
[Publications] H.Yamajo, H.Saruwatari, T.Takatani, T.Nishikawa, K.Shikano: "Blind Separation and Deconvolution for Convolutive Mixture of Speech Using SIMO-Model-Based ICA and Multichannel Inverse Filtering"Proceedings of 8th European Conference on Speech Communication and Technology. 537-540 (2003)
[Publications] S.Yoshizawa, K.Shikano: "Model-Integration Rapid Training based on Maximum Likelihood for Speech Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2601-2604 (2003)
[Publications] Y.Nakajima, H.Kashioka, N.Campbell, K.Shikano: "Non-Audible Murmur Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2601-2604 (2003)
[Publications] T.Nishiura, S.Nakamura, K.Miki, K.Shikano: "Environmental Sound Source Identification Based on Hidden Markov Model for Robust Speech Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2157-2160 (2003)
[Publications] H.Saruwatari, H.Yamajo, T.Takatani, T.Nishikawa, K.Shikano: "Blind Separation and Deconvolution of MIMO System Driven by Colored Inputs Using SIMO-Model-Based ICA with Information-Geometric Learning"Proceedings of IEEE Neural Network for Signal Processing Work shop 2003. 379-388 (2003)
[Publications] Hiroaki Yamajo, Hiroshi Saruwatari, Tomoya Takatani, Tsuyoki Nishikawa, Kiyohiro Shikano: "Evaluation of Blind Separation and Deconvolution for Convolutive Speech Mixture Using SIMO-Model-Based ICA"Proceedings of IWAENC2003. 299-302 (2003)
[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "High-Fildeity Blind Separation of Acoustic Signals using SIMO-Model-Based ICA with Information-Geometric Learning"Proceedings of IWAENC2003. 251-254 (2003)
[Publications] Raj Kishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Problems in Blind Separation of Convolutive Speech Mixtures by Negentropy Maximization"Proceedings of IWAENC2003. 287-290 (2003)
[Publications] Hiroshi Saruwatari, Hiroaki Yamajo, Tomoya Takatani, Tsuyoki Nishikawa, Kiyohiro Shikano: "Blind Separation and Deconvolution of MIMO-FIR System with Colored Sound Inputs Using SIMO-Model-Based ICA"Proc.2003 IEEE Workshop on Statistical Signal Processing. 421-424 (2003)
[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "SIMO-Model-Based ICA with Information-Geometric Learning for High-Fidelity Blind Source Separation"Proc.2003 International Symposium on Intelligent Signal Processing and Communication Systems. 108-113 (2003)
[Publications] P.Heracleous, Y.Nakajima, A.Lee, H.Saruwatari, K.Shikano: "Accurate Hidden Markov Models for Non-Audible Murmur (NAM) Recognition Based on Iterative Supervised Adaptation"IEEE Automatic Speech Recognition and Understanding Workshop. 73-76 (2003)
[Publications] Tsuyoki Nishikawa, Hiroshi Abe, Hiroshi Saruwatari, Kiyohiro Shikano: "Overdetermined Blind Source Separation of Real Acoustic Sounds Based on Multistage ICA Using Subarray Processing"Proc.2003 International Symposium on Signal Processing and Information Technology. (2003)
[Publications] H.Saruwatari, T.Takatani, H.Yamajo, K.Shikano: "SIMO-Model-Based Blind Acoustic Signal Separation : Concept and Its Application"Special Workshop in MAUI. (2004)
[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Combining Null beamformer and Fixed-point ICA for the Blind Separation of Convolutive Mixture of Speech"International conference on Intelligent Signal Processing and Robotics. (2004)
[Publications] Nobuo Kawaguchi, Kazuya Takeda, Fumitada Itakura: "MULTIMEDIA CORPUS OF IN-CAR SPEECH COMMUNICATION"Journal of VLSI Signal Processing Systems. Vol.36. (2004)
[Publications] 木村敏幸, 筧一彦, 武田一哉, 板倉文忠: "音場再現におけるマルチチャネル信号の空間符号化"日本バーチャルリアリティ学会論文誌. Vol.8, No.4. 433-442 (2003)
[Publications] Hiroshii Saruwatari, Satoshi Kurita, Kazuya Takeda, Fumitada Itakura, Tsuyoki Kiyohiro Shikano: "Blind source separation combining independent component analysis and beamforming"EURASIP Journal on Applied Signal Processing. Vol.2003, No.11. 1135-1146 (2003)
[Publications] 瀬川修, 武田一哉, 板倉文忠: "端点検出を行なわない連続音声認識"電気学会論文誌C. Vol.124-C. (2004)
[Publications] 荒木潤二, 西野隆典, 武田一哉, 板倉文忠: "スパーク音源を用いた頭部伝達関数の測定"日本音響学会誌,技術報告. (2004)
[Publications] Taisuke Ito, Kazuya Takeda, Fumitada Itakura: "Analysis and Recognition of Whispered Speech"Speech Communication. (2004)
[Publications] Hideki Banno, Tetsuya Shinde, Kazuya Takeda, Fumitada Itakura: "IN-CAR SPEECH RECOGNITION USING DISTRIBUTED MICROPHONES -ADAPTING TO AUTOMATICALLY DETECTED DRIVING CONDITIONS"Proc.of IEEE International Conference on Acoustics Speech and Signal Processing. (2003)
[Publications] Nobuo Kawaguchi, Shigeki Matsubara, Hiroya Murao, Kazuya Takeda, Fumitada Itakura: "CIAIR IN-CAR SPOKEN DIALOGUE CORPUS AND ITS APPLICATION"Proceedings in Special Workshop in MAUI. (2004)
[Publications] Toshihiro Isobe, Shoji Hayakawa, Hiroya Murao, Ryuji Mizutani, Kazuya Takeda: "A Study on Domain Recognition of Spoken Dialog System"Proc.of 8th European Conference on Speech Communication and Technology. (2003)
[Publications] Nobuo Kawaguchi, Shigeki Matsubara, Itsuki Kishida, Kazuya Takeda, Fumitada Itakura: "CONSTRUCTION AND ANALYSIS OF THE MULTI-LAYERED IN-CAR SPOKEN DIALOGUE CORPUS"The Workshop on DSP in Mobile and Vehicular Systems. (2003)
[Publications] 磯部俊洋, 早川昭二, 村尾浩也, 水谷龍治, 武田一哉, 板倉文忠: "複数モデルを選択的に用いる音声対話システムにおけるドメインの切り替え尺度の検討"情報処理学会研究報告. SLP-47. 41-46 (2003)
[Publications] 番弘光, 伊藤克亘, 武田一哉, 板倉文忠: "タッピングを利用した音声認識の検討"情報処理学会研究報告. SLP-47. 71-76 (2003)
[Publications] 五十嵐圭, 伊藤克亘, 武田一哉, 板倉文忠: "自動車運転動作からの個人性抽出に関する検討"電子情報通信学会基礎・境界ソサイエティ大会. A-17-1. 172 (2003)
[Publications] 番弘光, 伊藤克亘, 武田一哉, 板倉文忠: "タッピングによる単語境界の推定とその音声認識への利用に関する検討"日本音響学会2003年秋季研究発表会講演論文集. 1. 157-158 (2003)
[Publications] 内田和宏, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "聴覚マスキングを用いた頭外音像定位の評価法"日本音響学会講演論文集. 1-P-2. 581-582 (2003)
[Publications] 太田護, 伊藤克亘, 武田一哉, 板倉文忠: "言語モデルの対話文へのオンライン適応の検討"日本音響学会講演論文集. 3-Q-18. 161-162 (2003)
[Publications] 李衛鋒, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "SVMを用いたケプストラムの回帰による車内音声認識"日本音響学会講演論文集. 3-Q-7. 139-140 (2003)
[Publications] 西野隆典, 武田一哉, 板倉文忠: "クラスタリング手法による汎用HRTFの選択手法の検討"日本音響学会講演論文集. 2-5-4. 517-518 (2003)
[Publications] 瀬川修, 武田一哉, 板倉文忠: "言語モデルの予測方向が異なるサーチ結果の比較による信頼度判定方法"日本音響学会講演論文集. 155-156 (2003)
[Publications] 武田一哉, 河口信夫, アブット・フセイン, 板倉文忠: "音声・映像・運転行動コーパスを用いた車内情報インタフェースの研究"第2回情報科学技術フォーラム. 4-5 (2003)
[Publications] 内田和宏, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "聴覚マスキングを用いた頭外音像定位の評価法の検討"電子情報通信学会技術研究報告. EA2003-121. 15-20 (2003)
[Publications] 白勢彩子, 原直, 藤村浩司, 伊藤克亘, 武田一哉, 板倉文忠: "ユーザ評価と達成度との相関に基づく音声対話システムの品質評価の予備的検討"電子情報通信学会技術研究報告. NLC2003-92SP2003-155. 7-12 (2003)
[Publications] 藤村浩司, 伊藤克亘, 武田一哉, 板倉文忠: "大規模車内音声データベースにおける認識性能変動要因調査"電子情報通信学会技術研究報告. NLC2003-55SP2003-118. 31-36 (2003)
[Publications] 村尾浩也, 河口信夫, 松原茂樹, 山口由紀子, 武田一哉, 稲垣康善: "WOZによるオンライン修正が可能な事例ベース音声対話システム"情報処理学会研究報告. 2003-SLP-49. 269-274 (2003)
[Publications] 河口信夫, 松原茂樹, 山口由紀子, 武田一哉, 板倉文忠: "CIAIR実走行車内音声データベース"情報処理学会研究報告. 2003-SLP-49. 130-144 (2003)
[Publications] 木村敏幸, 寛一彦, 武田一哉, 板倉文忠: "波面合成による音場再現における制御点数の影響-方向感の検討-"日本音響学会講演論文集. 2-5-7. 523-524 (2003)
[Publications] M.Nishida, T.Kawahara: "Speaker model selection based on Bayesian information criterion applied to unsupervised speaker indexing"IEEE Trans. Speech & Audio Process. (2004)
[Publications] K.Komatani, S.Ueno, T.Kawahara, H.G.Okuno: "User modeling in spoken dialogue systems to generate flexible guidance"User Modeling and User-Adapted Interaction. (2004)
[Publications] H.Nanjo, T.Kawahara: "Language model and speaking rate adaptation for spontaneous presentation speech recognition"IEEE Trans. Speech & Audio Process. (2004)
[Publications] T.Kawahara, M.Hasegawa, K.Shitaoka, T.Kitade, H.Nanjo: "Automatic indexing of lecture presentations using unsupervised learning of presumed discourse markers"IEEE Trans. Speech & Audio Process. (2004)
[Publications] 南條浩輝, 河原達也: "講演音声認識のための教師なし言語モデル適応と発話速度に適応したデコーディング"電子情報通信学会論文誌. (2004)
[Publications] 下岡和也, 南條浩輝, 河原達也: "講演の書き起こしに対する統計的手法を用いた文体の整形"自然言語処理. (2004)
[Publications] 西田昌史, 河原達也: "BICに基づく統計的話者モデル選択による教師なし話者インデキシング"電子情報通信学会論文誌. Vol.J87-DII, No.2. 504-512 (2004)
[Publications] 秋田祐哉, 河原達也: "多数話者モデルを用いた討論音声の教師なし話者インデキシング"電子情報通信学会論文誌. Vol.J87-DII, No.2. 495-503 (2004)
[Publications] 山肩洋子, 河厚達也, 奥乃博, 美濃導彦: "音声対話システムにおける物体指示のための信念ネットワークを用いた曖昧性の解消"人工知能学会誌. Vol.19, No.1. 47-56 (2004)
[Publications] 駒谷和範, 鹿島博晶, 田中克明, 河原達也: "複合的言語制約に基づくキーフレーズ検出を用いた汎用的なデータベース検索音声対話プラットフォーム"情報処理学会論文誌. Vol.44, No.5. 1333-1342 (2003)
[Publications] 井本和範, 坪田康, 河原達也, 壇辻正剛: "英語韻律発音学習支援システムのための英語文強勢のモデル化と自動検出"日本音響学会誌. Vol.59, No.4. 183-191 (2003)
[Publications] 南條浩輝, 加藤一臣, 李晃伸, 河原達也: "大規模な日本語話し言葉データベースを用いた講演音声認識"電子情報通信学会論文誌. Vol.J86-DII, No.4. 450-459 (2003)
[Publications] Y.Tsubota, T.Kawahara, M.Dantsuji: "Formant structure estimation using vocal tract length normalization for CALL system"Acoustical Science & Technology. Vol.24, No.2. 93-96 (2003)
[Publications] 奥田浩三, 河原達也, 中村哲: "尤度基準による分析周期・窓長の自動選択手法を用いた発話速度の補正と音響モデルの構築"電子情報通信学会論文誌. Vol.J86-DII, No.2. 204-211 (2003)
[Publications] H.Nanjo, T.Kitade, T.Kawahara: "Automatic indexing of key sentences for lecture archives using statistics of presumed discourse markers"In Proc.IEEE-ICASSP. (2004)
[Publications] I.R.Lane, T.Kawahara, T.Matsui, S.Nakamura: "Out-of-domain detection based on confidence measures from multiple topic classification"In Proc.IEEE-ICASSP. (2004)
[Publications] N.Nishida, T.Kawahara: "Speaker indexing and adaptation using speaker clustering based on statistical model selection"In Proc.IEEE-ICASSP. (2004)
[Publications] A.Lee, K.Shikano, T.Kawahara: "Real-time word confidence scoring using local posterior probabilities on tree trellis search"In Proc.IEEE-ICASSP. (2004)
[Publications] K.Komatani, R.Ito, T.Kawahara, H.G.Okuno: "Recognition of emotional states in spoken dialogue with a robot"In Proc.Int'l Conf.Industrial & Engineering Applications of Artificial Intelligence & Expert Systems. (2004)
[Publications] T.Kawahara: "Automatic speech transcription and archiving system using the Corpus of Spontaneous Japanese"In Proc.International Congress on Acoustics (invited talk). (2004)
[Publications] T.Kawahara, T.Kitade, K.Shitaoka, H.Nanjo: "Efficient access to lecture audio archives through spoken language processing"In Proc.Special Workshop in Maui. (2004)
[Publications] T.Kawahara, K.Shitaoka, T.Kitade, H.Nanjo: "Automatic indexing of key sentences for lecture archives"In Proc.IEEE workshop on Automatic Speech Recognition and Understanding. (2003)
[Publications] Y.Tsubota, M.Dantsuji, T.Kawahara: "An English pronunciation learning system for Japanese students based on diagnosis of critical pronunciation errors"In Proc.EUROCALL. 204 (2003)
[Publications] A.Akita, T.Kawahara: "Unsupervised speaker indexing using anchor models and automatic transcription of discussions"In Proc.EUROSPEECH. 2985-2988 (2003)
[Publications] M.Nishida, T.Kawahara: "Speaker model selection using Bayesian information criterion for speaker indexing and speaker adaptation"In Proc.EUROSPEECH. 1849-1852 (2003)
[Publications] T.Kawahara, R.Ito, K.Komatani: "Spoken dialogue system for queries on appliance manuals using hierarchical confirmation strategy"In Proc.EUROSPEECH. 1701-1704 (2003)
[Publications] K.Komatani, S.Ueno, T.Kawahara, H.G.Okuno: "User modeling in spoken dialogue systems for flexible guidance generation"In Proc.EUROSPEECH. 745-748 (2003)
[Publications] I.R.Lane, T.Matsui, S.Nakamura, T.Kawahara: "Hierarchical topic classification for dialog speech recognition based on language model switching"In Proc.EUROSPEECH. (2003)
[Publications] K.Komatani, S.Ueno, T.Kawahara, H.G.Okuno: "Flexible guidance generation using user model in spoken dialogue systems"In Proc.Annual Meeting of Association for Computational Linguistics. 256-263 (2003)
[Publications] Y.Kiyota, S.Kurohashi, T.Misu, K.Komatani, T.Kawahara, F.Kido: "Dialog navigator : A spoken dialog Q-A system based on large text knowledge base"In Proc.ACL-03 Interactive Poster & Demo. 149-152 (2003)
[Publications] K.Komatani, F.Adachi, S.Ueno, T.Kawahara, H.G.Okuno: "Flexible spoken dialogue system based on user models and dynamic generation of VoiceXML scripts"In Proc.SIGdial Workshop Discourse & Dialogue. 87-96 (2003)
[Publications] T.Kawahara, H.Nanjo, T.Shinozaki, S.Furui: "Benchmark test for speech recognition using the Corpus of Spontaneous Japanese"In Proc.ISCA & IEEE Workshop on Spontaneous Speech Processing and Recognition. 135-138 (2003)
[Publications] H.Nanjo, K.Shitaoka, T.Kawahara: "Automatic transformation of lecture transcription into document style using statistical framework"In Proc.ISCA & IEEE Workshop on Spontaneous Speech Processing and Recognition. 215-218 (2003)

2003 Fiscal Year Annual Research Report

分散して蓄積された音声データを用いて多様な音響モデルを作成する方法の研究

Principal Investigator

武田 一哉 名古屋大学, 情報科学研究科, 教授 (20273295)

Research Products

[Publications] T.Nishikawa, H.Saruwatari, K.Shikano: "Blind source separation of acoustic signals based on multistage ICA combining frequency-domain ICA and time-domain ICA"IEICE Trans.Fundamentals. Vol.E86-A, No.4. 846-858 (2003)

[Publications] 南條 浩輝, 加藤 一臣, 李 晃伸, 河原 達也: "大規模な日本語話し言葉データベースを用いた講演音声認識"電子情報通信学会論文誌. Vol.J86-D-II No.4. 450-459 (2003)

[Publications] T.Nishikawa, H.Saruwatari, K.Shikano: "Stable learning algorithm for blind separation of temporally correlated acoustic signals combining multistage ICA and Linear Prediction"IEICE Trans.Fundamentals. Vol.E86-A, No.8. 2028-2036 (2003)

[Publications] Panikos Heracleous, Satoshi Nakamura, Kiyohiro Shikano: "Simultaneous Recognition of Distant-Talking Speech of Multiple Talkers Based on the 3-D N-best Search"Journal of VLSI Signal Processing, Kluwer Academic Publishers. Vol.36-2. 105-116 (2004)

[Publications] H.Saruwatari, S.Kurita, K.Takeda, F.Itakura, T.Nishikawa.K.Shikano: "Blind Source Separation Combining Independent Component Analysis and Beamforming"EURASIP Journal on Applied Signal Processing. Vol.2003, No.11. 1135-1146 (2003)

[Publications] S.Yamade, A.Lee, H.Saruwatari, K.Shikano: "Unsupervised Speaker Adaptation Based on HMM Sufficient Statistics in Various Noisy Environments"Proceedings of 8th European Conference on Speech Communication and Technology. 1493-1496 (2003)

[Publications] 舛田剛志, 戸田智基, 川波弘道, 猿渡洋, 鹿野清宏: "韻律的に多重化した音声データベースの設計と発話速度におけるその評価"電子情報通信学会論文誌D-II. No.2. 447-455 (2004)

[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Probability Distribution of Time-Series of Speech Spectral Components"IEICE Trans.Fundamentals. Vol.E87-A, No.3. 584-597 (2004)

[Publications] 永田悠, 立蔵洋介, 猿渡洋, 鹿野清宏: "音場再現システムにおける環境変化に適応的な逆フィルターの逐次的緩和アルゴリズム"電子情報通信学会論文誌A. Vol.J86-A, No.8. 824-834 (2003)

[Publications] 西村竜一, 西原洋平, 鶴身玲典, 李晃伸, 猿渡洋, 鹿野清宏: "実環境研究プラットホームとしての音声情報案内システムの運用"電子情報通信学会論文誌. Vol.J87-D-II No.3. 789-798 (2004)

[Publications] Y.Hinamoto, K.Mino, H.Saruwatari, K.Shikano: "Spoken dialogue system based on sound field control and microphone array"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.V. 505-508 (2003)

[Publications] T.Shiraishi, T.Toda, H.Kawanami, H.Saruwatari, K.Shikano: "Simple Designing Methods of Corpus-Based Visual Speech Synthesis"Proceedings of 8th European Conference on Speech Communication and Technology. 2241-2244 (2003)

[Publications] T.Toda, H.Kawai, M.Tsuzaki, K.Shikano: "SEGMENT SELECTION CONSIDERING LOCAL DEGRADATION OF NATURALNESS IN CONCATENATIVE SPEECH SYNTHESIS"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.I. 696-699 (2003)

[Publications] H.Kawanami, Y.Iwami, T.Toda, H.Saruwatari, K.Shikano: "GMM-based Voice Conversion Applied to Emotional Speech Synthesis"Proceedings of 8th European Conference on Speech Communication and Technology. 2401-2404 (2003)

[Publications] Y.Nakajima, H.Kashino, K.Shikano, N.Campbell: "RECOGNITION INPUT INTERFACE USING STETHOSCOPIC MICROPHONE ATTACHED TO THE SKIN"Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing. Vol.V. 708-711 (2003)

[Publications] P.Heracleous, S.Nakamura, K.Shikano: "A Semi-Blind Source Separation Method for Hands-Free Speech Recognition of Multiple Talkers"Proceedings of 8th European Conference on Speech Communication and Technology. 509-512 (2003)

[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Akinobu Lee, Kiyohiro Shikano: "A Fixed-Point ICA Algorithm for Convoluted Speech Signal Separation"Proceedings of Fourth International Symposium on Independent Component Analysis and Blind Signal Separation. 579-584 (2003)

[Publications] R.Nisimura, Y.Nishihara, R.Tsurumi, A.Lee, H.Saruwatari, K.Shikano: "Takemaru-kun : Speech-Oriented Information System for Real World Research Platform"International Workshop on Language Understanding and Agents for Real World Interaction. 70-78 (2003)

[Publications] S.Yoshizawa, K.Shikano: "Model-Integration Rapid Training based on Maximum Likelihood for Speech Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2601-2604 (2003)

[Publications] Y.Nakajima, H.Kashioka, N.Campbell, K.Shikano: "Non-Audible Murmur Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2601-2604 (2003)

[Publications] T.Nishiura, S.Nakamura, K.Miki, K.Shikano: "Environmental Sound Source Identification Based on Hidden Markov Model for Robust Speech Recognition"Proceedings of 8th European Conference on Speech Communication and Technology. 2157-2160 (2003)

[Publications] Hiroaki Yamajo, Hiroshi Saruwatari, Tomoya Takatani, Tsuyoki Nishikawa, Kiyohiro Shikano: "Evaluation of Blind Separation and Deconvolution for Convolutive Speech Mixture Using SIMO-Model-Based ICA"Proceedings of IWAENC2003. 299-302 (2003)

[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "High-Fildeity Blind Separation of Acoustic Signals using SIMO-Model-Based ICA with Information-Geometric Learning"Proceedings of IWAENC2003. 251-254 (2003)

[Publications] Raj Kishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Problems in Blind Separation of Convolutive Speech Mixtures by Negentropy Maximization"Proceedings of IWAENC2003. 287-290 (2003)

[Publications] Hiroshi Saruwatari, Hiroaki Yamajo, Tomoya Takatani, Tsuyoki Nishikawa, Kiyohiro Shikano: "Blind Separation and Deconvolution of MIMO-FIR System with Colored Sound Inputs Using SIMO-Model-Based ICA"Proc.2003 IEEE Workshop on Statistical Signal Processing. 421-424 (2003)

[Publications] Tomoya Takatani, Tsuyoki Nishikawa, Hiroshi Saruwatari, Kiyohiro Shikano: "SIMO-Model-Based ICA with Information-Geometric Learning for High-Fidelity Blind Source Separation"Proc.2003 International Symposium on Intelligent Signal Processing and Communication Systems. 108-113 (2003)

[Publications] P.Heracleous, Y.Nakajima, A.Lee, H.Saruwatari, K.Shikano: "Accurate Hidden Markov Models for Non-Audible Murmur (NAM) Recognition Based on Iterative Supervised Adaptation"IEEE Automatic Speech Recognition and Understanding Workshop. 73-76 (2003)

[Publications] Tsuyoki Nishikawa, Hiroshi Abe, Hiroshi Saruwatari, Kiyohiro Shikano: "Overdetermined Blind Source Separation of Real Acoustic Sounds Based on Multistage ICA Using Subarray Processing"Proc.2003 International Symposium on Signal Processing and Information Technology. (2003)

[Publications] H.Saruwatari, T.Takatani, H.Yamajo, K.Shikano: "SIMO-Model-Based Blind Acoustic Signal Separation : Concept and Its Application"Special Workshop in MAUI. (2004)

[Publications] Rajkishore Prasad, Hiroshi Saruwatari, Kiyohiro Shikano: "Combining Null beamformer and Fixed-point ICA for the Blind Separation of Convolutive Mixture of Speech"International conference on Intelligent Signal Processing and Robotics. (2004)

[Publications] Nobuo Kawaguchi, Kazuya Takeda, Fumitada Itakura: "MULTIMEDIA CORPUS OF IN-CAR SPEECH COMMUNICATION"Journal of VLSI Signal Processing Systems. Vol.36. (2004)

[Publications] 木村敏幸, 筧一彦, 武田一哉, 板倉文忠: "音場再現におけるマルチチャネル信号の空間符号化"日本バーチャルリアリティ学会論文誌. Vol.8, No.4. 433-442 (2003)

[Publications] Hiroshii Saruwatari, Satoshi Kurita, Kazuya Takeda, Fumitada Itakura, Tsuyoki Kiyohiro Shikano: "Blind source separation combining independent component analysis and beamforming"EURASIP Journal on Applied Signal Processing. Vol.2003, No.11. 1135-1146 (2003)

[Publications] 瀬川修, 武田一哉, 板倉文忠: "端点検出を行なわない連続音声認識"電気学会論文誌C. Vol.124-C. (2004)

[Publications] 荒木潤二, 西野隆典, 武田一哉, 板倉文忠: "スパーク音源を用いた頭部伝達関数の測定"日本音響学会誌,技術報告. (2004)

[Publications] Taisuke Ito, Kazuya Takeda, Fumitada Itakura: "Analysis and Recognition of Whispered Speech"Speech Communication. (2004)

[Publications] Hideki Banno, Tetsuya Shinde, Kazuya Takeda, Fumitada Itakura: "IN-CAR SPEECH RECOGNITION USING DISTRIBUTED MICROPHONES -ADAPTING TO AUTOMATICALLY DETECTED DRIVING CONDITIONS"Proc.of IEEE International Conference on Acoustics Speech and Signal Processing. (2003)

[Publications] Nobuo Kawaguchi, Shigeki Matsubara, Hiroya Murao, Kazuya Takeda, Fumitada Itakura: "CIAIR IN-CAR SPOKEN DIALOGUE CORPUS AND ITS APPLICATION"Proceedings in Special Workshop in MAUI. (2004)

[Publications] Toshihiro Isobe, Shoji Hayakawa, Hiroya Murao, Ryuji Mizutani, Kazuya Takeda: "A Study on Domain Recognition of Spoken Dialog System"Proc.of 8th European Conference on Speech Communication and Technology. (2003)

[Publications] Nobuo Kawaguchi, Shigeki Matsubara, Itsuki Kishida, Kazuya Takeda, Fumitada Itakura: "CONSTRUCTION AND ANALYSIS OF THE MULTI-LAYERED IN-CAR SPOKEN DIALOGUE CORPUS"The Workshop on DSP in Mobile and Vehicular Systems. (2003)

[Publications] 磯部俊洋, 早川昭二, 村尾浩也, 水谷龍治, 武田一哉, 板倉文忠: "複数モデルを選択的に用いる音声対話システムにおけるドメインの切り替え尺度の検討"情報処理学会研究報告. SLP-47. 41-46 (2003)

[Publications] 番弘光, 伊藤克亘, 武田一哉, 板倉文忠: "タッピングを利用した音声認識の検討"情報処理学会研究報告. SLP-47. 71-76 (2003)

[Publications] 五十嵐圭, 伊藤克亘, 武田一哉, 板倉文忠: "自動車運転動作からの個人性抽出に関する検討"電子情報通信学会基礎・境界ソサイエティ大会. A-17-1. 172 (2003)

[Publications] 番弘光, 伊藤克亘, 武田一哉, 板倉文忠: "タッピングによる単語境界の推定とその音声認識への利用に関する検討"日本音響学会2003年秋季研究発表会講演論文集. 1. 157-158 (2003)

[Publications] 内田和宏, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "聴覚マスキングを用いた頭外音像定位の評価法"日本音響学会講演論文集. 1-P-2. 581-582 (2003)

[Publications] 太田護, 伊藤克亘, 武田一哉, 板倉文忠: "言語モデルの対話文へのオンライン適応の検討"日本音響学会講演論文集. 3-Q-18. 161-162 (2003)

[Publications] 李衛鋒, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "SVMを用いたケプストラムの回帰による車内音声認識"日本音響学会講演論文集. 3-Q-7. 139-140 (2003)

[Publications] 西野隆典, 武田一哉, 板倉文忠: "クラスタリング手法による汎用HRTFの選択手法の検討"日本音響学会講演論文集. 2-5-4. 517-518 (2003)

[Publications] 瀬川 修, 武田一哉, 板倉文忠: "言語モデルの予測方向が異なるサーチ結果の比較による信頼度判定方法"日本音響学会講演論文集. 155-156 (2003)

[Publications] 武田一哉, 河口信夫, アブット・フセイン, 板倉文忠: "音声・映像・運転行動コーパスを用いた車内情報インタフェースの研究"第2回情報科学技術フォーラム. 4-5 (2003)

[Publications] 内田和宏, 西野隆典, 伊藤克亘, 武田一哉, 板倉文忠: "聴覚マスキングを用いた頭外音像定位の評価法の検討"電子情報通信学会技術研究報告. EA2003-121. 15-20 (2003)

[Publications] 白勢彩子, 原直, 藤村浩司, 伊藤克亘, 武田一哉, 板倉文忠: "ユーザ評価と達成度との相関に基づく音声対話システムの品質評価の予備的検討"電子情報通信学会技術研究報告. NLC2003-92SP2003-155. 7-12 (2003)

[Publications] 藤村浩司, 伊藤克亘, 武田一哉, 板倉文忠: "大規模車内音声データベースにおける認識性能変動要因調査"電子情報通信学会技術研究報告. NLC2003-55SP2003-118. 31-36 (2003)

[Publications] 村尾浩也, 河口信夫, 松原茂樹, 山口由紀子, 武田一哉, 稲垣康善: "WOZによるオンライン修正が可能な事例ベース音声対話システム"情報処理学会研究報告. 2003-SLP-49. 269-274 (2003)

[Publications] 河口信夫, 松原茂樹, 山口由紀子, 武田一哉, 板倉文忠: "CIAIR実走行車内音声データベース"情報処理学会研究報告. 2003-SLP-49. 130-144 (2003)

[Publications] 木村敏幸, 寛一彦, 武田一哉, 板倉文忠: "波面合成による音場再現における制御点数の影響-方向感の検討-"日本音響学会講演論文集. 2-5-7. 523-524 (2003)

[Publications] M.Nishida, T.Kawahara: "Speaker model selection based on Bayesian information criterion applied to unsupervised speaker indexing"IEEE Trans. Speech & Audio Process. (2004)

[Publications] K.Komatani, S.Ueno, T.Kawahara, H.G.Okuno: "User modeling in spoken dialogue systems to generate flexible guidance"User Modeling and User-Adapted Interaction. (2004)

[Publications] H.Nanjo, T.Kawahara: "Language model and speaking rate adaptation for spontaneous presentation speech recognition"IEEE Trans. Speech & Audio Process. (2004)

[Publications] T.Kawahara, M.Hasegawa, K.Shitaoka, T.Kitade, H.Nanjo: "Automatic indexing of lecture presentations using unsupervised learning of presumed discourse markers"IEEE Trans. Speech & Audio Process. (2004)

[Publications] 南條浩輝, 河原達也: "講演音声認識のための教師なし言語モデル適応と発話速度に適応したデコーディング"電子情報通信学会論文誌. (2004)

[Publications] 下岡和也, 南條浩輝, 河原達也: "講演の書き起こしに対する統計的手法を用いた文体の整形"自然言語処理. (2004)

[Publications] 西田昌史, 河原達也: "BICに基づく統計的話者モデル選択による教師なし話者インデキシング"電子情報通信学会論文誌. Vol.J87-DII, No.2. 504-512 (2004)

[Publications] 秋田祐哉, 河原達也: "多数話者モデルを用いた討論音声の教師なし話者インデキシング"電子情報通信学会論文誌. Vol.J87-DII, No.2. 495-503 (2004)

[Publications] 山肩洋子, 河厚達也, 奥乃博, 美濃導彦: "音声対話システムにおける物体指示のための信念ネットワークを用いた曖昧性の解消"人工知能学会誌. Vol.19, No.1. 47-56 (2004)

[Publications] 駒谷和範, 鹿島博晶, 田中克明, 河原達也: "複合的言語制約に基づくキーフレーズ検出を用いた汎用的なデータベース検索音声対話プラットフォーム"情報処理学会論文誌. Vol.44, No.5. 1333-1342 (2003)

[Publications] 井本和範, 坪田康, 河原達也, 壇辻正剛: "英語韻律発音学習支援システムのための英語文強勢のモデル化と自動検出"日本音響学会誌. Vol.59, No.4. 183-191 (2003)

[Publications] 南條浩輝, 加藤一臣, 李晃伸, 河原達也: "大規模な日本語話し言葉データベースを用いた講演音声認識"電子情報通信学会論文誌. Vol.J86-DII, No.4. 450-459 (2003)

[Publications] Y.Tsubota, T.Kawahara, M.Dantsuji: "Formant structure estimation using vocal tract length normalization for CALL system"Acoustical Science & Technology. Vol.24, No.2. 93-96 (2003)

[Publications] 奥田浩三, 河原達也, 中村哲: "尤度基準による分析周期・窓長の自動選択手法を用いた発話速度の補正と音響モデルの構築"電子情報通信学会論文誌. Vol.J86-DII, No.2. 204-211 (2003)

[Publications] H.Nanjo, T.Kitade, T.Kawahara: "Automatic indexing of key sentences for lecture archives using statistics of presumed discourse markers"In Proc.IEEE-ICASSP. (2004)

[Publications] I.R.Lane, T.Kawahara, T.Matsui, S.Nakamura: "Out-of-domain detection based on confidence measures from multiple topic classification"In Proc.IEEE-ICASSP. (2004)

[Publications] N.Nishida, T.Kawahara: "Speaker indexing and adaptation using speaker clustering based on statistical model selection"In Proc.IEEE-ICASSP. (2004)

[Publications] A.Lee, K.Shikano, T.Kawahara: "Real-time word confidence scoring using local posterior probabilities on tree trellis search"In Proc.IEEE-ICASSP. (2004)

[Publications] K.Komatani, R.Ito, T.Kawahara, H.G.Okuno: "Recognition of emotional states in spoken dialogue with a robot"In Proc.Int'l Conf.Industrial & Engineering Applications of Artificial Intelligence & Expert Systems. (2004)

武田一哉名古屋大学, 情報科学研究科, 教授 (20273295)

[Publications] 南條浩輝, 加藤一臣, 李晃伸, 河原達也: "大規模な日本語話し言葉データベースを用いた講演音声認識"電子情報通信学会論文誌. Vol.J86-D-II No.4. 450-459 (2003)

[Publications] 瀬川修, 武田一哉, 板倉文忠: "言語モデルの予測方向が異なるサーチ結果の比較による信頼度判定方法"日本音響学会講演論文集. 155-156 (2003)