• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to project page

2010 Fiscal Year Annual Research Report

新しい音声メディアによるユニバーサルコミュニケーションの研究

Research Project

Project/Area Number 19200009
Research InstitutionNara Institute of Science and Technology

Principal Investigator

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)

Co-Investigator(Kenkyū-buntansha) 猿渡 洋  奈良先端科学技術大学院大学, 情報科学研究科, 准教授 (30324974)
戸田 智基  奈良先端科学技術大学院大学, 情報科学研究科, 助教 (90403328)
川波 弘道  奈良先端科学技術大学院大学, 情報科学研究科, 助教 (80335489)
Keywords音声ユニバーサルインタフェース / 非可聴つぶやき(NAM) / ブラインド音源分離(BSS) / ハンズフリー音声認識
Research Abstract

引き続き(1)新しい静かな音声メディアである非可聴つぶやき(NAM)による無音声認識と無音声電話の研究、(2)歪みなしの音の分離技術SIMO-ICAを用いたハンズフリー音声認識システムの研究、(3)実環境での音声対話システムの研究を推進した。
(1)非可聴つぶやき(NAM)による音声コミュニケーション手段の研究
(a)NAMマイクの特性や話者の発話スタイルに、少ない任意の発話で適応できる無音声電話の構築を目指し、装着場所や発話スタイルの適応で成果をあげた。
(b)大量の通常発話も利用した話者変換手法を用いて、不特定話者のNAM音韻モデルを構築して、不特定話者NAM音声認識の性能を向上させた。
(2)歪なし音源分離SIMO-ICAによる音声コミュニケーション手段の研究
(a)種々の音環境でのミュージカルノイズの低減手法について、理論的な評価尺度の研究を進め、音環境に適応してミュージカルノイズの量を制御できるアルゴリズムを確立した。
(3)実環境音声対話システムの研究
(a)音声情報案内システムでのタスク外発話の検出能力の向上と、WebのVoice Searchの検出能力を向上させ、タスク外発話にも応答できるシステムを構築を行った。
(b)平城遷都1300年祭で4ヶ月間の運用して、音声情報案内システムのポータビリティ、Voice Searchの有効性などの評価を行った。

  • Research Products

    (23 results)

All 2010 Other

All Journal Article (5 results) (of which Peer Reviewed: 5 results) Presentation (17 results) Remarks (1 results)

  • [Journal Article] Silent-speech enhancement using body-conducted vocal-tract resonance signals2010

    • Author(s)
      Tatsuya Hirahara, Makoto Otani, Shota Shimizu, Tomoki Toda, Keigo Nakamura, Yoshitaka Nakajima, Kiyohiro Shikano
    • Journal Title

      Speech Communication

      Volume: vol.52, no.4 Pages: 301-313

    • Peer Reviewed
  • [Journal Article] Adaptive training for voice conversion based on eigenvoices2010

    • Author(s)
      Yamato Ohtani, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Journal Title

      IEICE Trans.Information and Systems

      Volume: vol.E93-D, no.6 Pages: 1589-1598

    • Peer Reviewed
  • [Journal Article] Musicalnoise analysis in methods of integrating microphone array and spectral subtraction based on higher-order statistics2010

    • Author(s)
      Yu Takahashi, Hiroshi Saruwatari, Kiyohiro Shikano, Kazunobu Kondo
    • Journal Title

      EURASIP Journal on Advances in Signal Processing

      Volume: vol.2010, Article ID 431347 Pages: 25

    • Peer Reviewed
  • [Journal Article] Evaluation of extremely small sound source signals used in speaking-aid system with statistical voice conversion2010

    • Author(s)
      Keigo Nakamura, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Journal Title

      IEICE Trans.Information and Systems

      Volume: vol.E93-D, no.7 Pages: 1909-1917

    • Peer Reviewed
  • [Journal Article] Esophageal speech enhancement based on statistical voice conversion with gaussian mixture models2010

    • Author(s)
      Hironori Doi, Keigo Nakamura, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Journal Title

      IEICE Trans.Information and Systems

      Volume: vol.E93-D, no.9 Pages: 2491-2499

    • Peer Reviewed
  • [Presentation] Acoustic Compensation Method for Accepting Different Recording Devices in Body-Conducted Voice Conversion2010

    • Author(s)
      Daisuke Deguchi, Hironori Doi, Tomoki 'Ibda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of APSIPA Annual Summit and Conference 2010 (APSIPA2010)
    • Place of Presentation
      Biopolis, Singapore
    • Year and Date
      2010-12-16
  • [Presentation] Speaking-Aid Systems Based on One-to-Many Eigenvoice Conversion for Total Laryngectomees2010

    • Author(s)
      Hironori Doi, Keigo Nakamura, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of APSIPA Annual Summit and Conference 2010 (APSIPA2010)
    • Place of Presentation
      Biopolis, Singapore
    • Year and Date
      2010-12-16
  • [Presentation] Training Data Size Requirements for Topic Classification in a Speech-Oriented Guidance System2010

    • Author(s)
      Rafael Torres, Hiromichi Kawanami, Thmoko Matsui, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of APSIPA Annual Summit and Conference 2010 (APSIPA2010)
    • Place of Presentation
      Biopolis, Singapore
    • Year and Date
      2010-12-16
  • [Presentation] Theoretical analysis of musicial noise in wiener filter via higher-order statistics2010

    • Author(s)
      Takayuki Inoue, Hiroshi Saruwatari, Kiyohiro Shikano, Kazunobu Kondo
    • Organizer
      Proc.of APSIPA Annual Summit and Conference 2010 (APSIPA2010)
    • Place of Presentation
      Biopolis, Singapore
    • Year and Date
      2010-12-15
  • [Presentation] An evaluation of discriminative training for hidden Markov models in a real-environment speech-oriented guidance system2010

    • Author(s)
      Denis Babani, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of APSIPA Annual Summit and Conference 2010 (APSIPA2010)
    • Place of Presentation
      Biopolis, Singapore
    • Year and Date
      2010-12-14
  • [Presentation] Blind speech extraction combining ICA-based noise estimation and less-musical-noise nonlinear post processing2010

    • Author(s)
      Hiroshi Saruwatari, Yu Takahashi, Kiyohiro Shikano, Kazunobu Kondo
    • Organizer
      Asilomar Conference on Signals, Systems, and Computers
    • Place of Presentation
      Asilomar, California, USA
    • Year and Date
      2010-11-09
  • [Presentation] Improvement of speech recognition performance for spoken-oriented robot dialog system using end-fire array2010

    • Author(s)
      Hiroshi Sawada, Jani Even, Hiroshi Saruwatari, Kiyohiro Shikano, Tomoya Takatani
    • Organizer
      Proc.of 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2010)
    • Place of Presentation
      Taipei, Taiwan
    • Year and Date
      2010-10-19
  • [Presentation] The use of air-pressure sensor in electrolaryngeal speech enhancement based on statistical voice conversion2010

    • Author(s)
      Keigo Nakamura, Tomoki Toda, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of INTERSPEECH
    • Place of Presentation
      Chiba City, Japan
    • Year and Date
      2010-09-29
  • [Presentation] Adaptive voice-quality control based on one-to-many eigenvoice conversion2010

    • Author(s)
      Kumi Ohta, Tomoki Toda, Yama to Ohtani, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of INTERSPEECH
    • Place of Presentation
      Chiba City, Japan
    • Year and Date
      2010-09-29
  • [Presentation] Comparison of Methods for Topic Classification in a Speech-Oriented Guidance System2010

    • Author(s)
      Rafael Torres, Shota Takeuchi, Hiromichi Kawanami, Tomoko Matsui, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of INTERSPEECH
    • Place of Presentation
      Chiba City, Japan
    • Year and Date
      2010-09-28
  • [Presentation] Blind speech extraction combining generalized MMSE STSA estimator and ICA-based noise and speech probability density function estimations2010

    • Author(s)
      Hiroshi Saruwatari, Ryoi Okamoto, Yu Takahashi, Kiyohiro Shikano
    • Organizer
      Proc.of the 9th International Conference on Latent Variable Analysis and Signal Separation (LVA2010)
    • Place of Presentation
      St.Malo, France
    • Year and Date
      2010-09-27
  • [Presentation] Blind signal extraction based joint suppression of diffuse background noise and late reverberation2010

    • Author(s)
      Jani Even, Hiroshi Saruwatari, Kiyohiro Shikano, Tomoya Takatani
    • Organizer
      Proc.of European Signal Processing Conference (EUSIPCO2010)
    • Place of Presentation
      Aalborg, Denmark
    • Year and Date
      2010-09-26
  • [Presentation] Linear transformation approaches to many-to-one voice conversion2010

    • Author(s)
      Chie Hayashida, Tomoki Toda, Yamato Ohtani, Hiroshi Saruwatari, Kiyohiro Shikano
    • Organizer
      Proc.of the 7th ISCA Speech Synthesis Workshop (SSW7)
    • Place of Presentation
      Kyoto City, Japan
    • Year and Date
      2010-09-22
  • [Presentation] Theoretical analysis of iterative weak spectral subtraction via higher-order statistics2010

    • Author(s)
      Takayuki Inoue, Hiroshi Saruwatari, Yu Takahashi, Kiyohiro Shikano, Kazunobu Kondo
    • Organizer
      Proc.of 2010 IEEE International Workshop on Machine Learning for Signal Processing (MLSP2010)
    • Place of Presentation
      Kittila, Finland
    • Year and Date
      2010-09-01
  • [Presentation] Binaural hearing aid using sound-localization-preserved MMSE STSA estimator with ICA-based noise estimation2010

    • Author(s)
      Hiroshi Saruwatari, Masanobu Go, Ryoi Okamoto, Kiyohiro Shikano
    • Organizer
      Proc.of International Workshop on Acoustic Echo and Noise Control (IWAENC2010)
    • Place of Presentation
      Tel Aviv, Israel
    • Year and Date
      2010-08-30
  • [Presentation] Theoretical analysis of musical noise in generalized spectral subtraction : why should not use power/amplitude subtraction?2010

    • Author(s)
      Takayuki Inoue, Yu Takahashi, Hiroshi Saruwatari, Kiyohiro Shikano, Kazunobu Kondo
    • Organizer
      Proc.of European Signal Processing Conference (EUSIPCO2010)
    • Place of Presentation
      Aalborg, Denmark
    • Year and Date
      2010-08-26
  • [Presentation] Musical noise controllable algorithm of channelwise spectral subtraction and beamforming based on higher-order statistics criterion2010

    • Author(s)
      Yohei Ishikawa, Hiroshi Saruwatari, Yu Takahashi, Kiyohiro Shikano, Kazunobu Kondo
    • Organizer
      Proc.of the 2nd International Workshop on Cognitive Information Processing (CIP2010)
    • Place of Presentation
      Elba, Italy
    • Year and Date
      2010-06-14
  • [Remarks]

    • URL

      http://spalab.naist.jp/database/library/paper_10.html

URL: 

Published: 2012-07-19  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi