1990 Fiscal Year Annual Research Report

音声情報処理研究用日本語音声デ-タベ-スの作成

Research Project

Project/Area Number	01850068
Research Category	Grant-in-Aid for Developmental Scientific Research (B)
Research Institution	University of Tsukuba
Principal Investigator	板橋秀一筑波大学, 電子・情報工学系, 教授 (70151454)
Co-Investigator(Kenkyū-buntansha)	牧野正三東北大学, 応用情報学研究センター, 助教授 (00089806) 小畑秀文東京農工大学, 工学部, 教授 (80013720) 白井克彦早稲田大学, 理工学部, 教授 (10063702) 藤崎博也東京大学, 工学部, 教授 (80010776)
Keywords	音声情報処理 / 音声デ-タベ-ス / 日本語 / 音声
Research Abstract	DAT音声デ-タの編集:初年度に収録した男女各10名計20名の話者による音声デ-タ(各項目4回発声、1名当り2時間計40時間余りの音声)を検聴し、録音・発声状態の良好なもの、男女各6名計12名の音声デ-タ(1名当たり約2時間計24時間)を選んだ。それらについて、検索が便利になるように各項目ごとにプログラム番号をつけて、音声デ-タベ-ス用マスタ-テ-プを作成した。また、各発声項目の発音の状態や録音状況を詳細に記述した検聴表を作成した。 CDーROMの作成:上記DATデ-タの中、連続音声資料7種について、4回の発声中最も品質の良いもの1回を選び、それをCDーROMに記録した。これは日本語の文章音声のCDーROM版音声デ-タベ-スとしては日本で初めての試みといって良いであろう。音声デ-タベ-スの主要な目的の一つは、音声分析・合成や音声認識手法の開発およびその比較・評価に利用することである。そこで、種々の方法による音声分析・認識実験を試みた。騒音中の音声認識については、定常・非定常騒音が混在している場合でも、非定常騒音が音声と重なっていない場合、最尤法による音声強調によってS/N比が13dB改善されることを示した。

Research Products
(15 results)

All Other

All Publications (15 results)

[Publications] Shuichi ITAHASHI: "Recent Speech Database Projects in Japan" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 24.1. 1081-1084 (1990)
[Publications] Shichi ITAHASHI: "On Speech Database Efforts in Japan" Preprints of International Symposium on International Coordination and Standardization of Speech Database and Assessment Techniques for Speech Input/Output. 57-63 (1990)
[Publications] Jingxu CUI: "A Comparison of the Articulation of the Chinese/i,,l/by Chinese and Japanese Speakers" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 15.10. 629-632 (1990)
[Publications] Hiroya FUJISAKI and S.SATO: "Proposal and Evaluation of a New Scheme for Reliable Pitch Extraction of Speech" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 11.14. 473-476 (1990)
[Publications] Hiroya FUJISAKI: "Influence of context and knowkedge on the Perception of Continuous Speech" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 10.9. 417-420 (1990)
[Publications] Katsuhiko SHIRAI: "Speaker Adaptable Phoneme Recognition Selecting Reliable Acoustic Features based on Mutual Information" Proc.of ICSLP90ーInternational conference on Spoken Language Processing. 353-356 (1990)
[Publications] Katsuhiko SHIRAI: "Speech Synthesis Using Superposition of Sinusoidal Waves Generated by Synchronized Oscillators" Proc.of ICSLP90ーInternational conference on Spoken Language Processing. 345-348 (1990)
[Publications] K.GYOUTOKU,(H.KOBATAKE): "Maximum likelihood estimation of speech waveform under nonstationary noise environments" Proc.ICSLP90. 1149-1152 (1990)
[Publications] Hidefumi KOBATAKE: "Enhancement of noisy speech by maximum likelihood estimation" Proc.ICASSP90. (1991)
[Publications] 城風敏彦、(牧野正三): "発話全体の連続性を考慮した基本周波数の抽出" 電子情報通信学会論文誌. J73ーA. 1537-1539 (1990)
[Publications] 松尾広、(牧野正三): "音素の持続時間モデルに基づく検証法を用いた単語音声認識" 電子情報通信学会論文誌. J73ーDーII. 1936-1944 (1990)
[Publications] 城風敏彦、(牧野正三) 城戸健一: "音声情報の自動獲得機能を持つ分散型大規模音声デ-タベ-ス「KーDB」" 情報処理学会論文誌. 32. 62-70 (1991)
[Publications] 榑松明: "ATR Japanese Speech Database as a tool of Speech Recognition and Synthesis" SPEECH COMMUNICATION. Volume 9 No.4. 357-357 (1990)
[Publications] 榑松明: "A Perspective of telephone interpretation research" Proceedign of Pacific Rim International Conference on Artificial Intelligence. 11-16 (1990)
[Publications] 森元逞(榑松明): "Spoken language translation toward realizing an automatic telephone interpretation system" Proceeding of InfoーJapan'90(情報処理学会30周年記念国際会議). 553-560 (1990)

1990 Fiscal Year Annual Research Report

音声情報処理研究用日本語音声デ-タベ-スの作成

Principal Investigator

板橋 秀一 筑波大学, 電子・情報工学系, 教授 (70151454)

Research Products

[Publications] Shuichi ITAHASHI: "Recent Speech Database Projects in Japan" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 24.1. 1081-1084 (1990)

[Publications] Shichi ITAHASHI: "On Speech Database Efforts in Japan" Preprints of International Symposium on International Coordination and Standardization of Speech Database and Assessment Techniques for Speech Input/Output. 57-63 (1990)

[Publications] Jingxu CUI: "A Comparison of the Articulation of the Chinese/i,,l/by Chinese and Japanese Speakers" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 15.10. 629-632 (1990)

[Publications] Hiroya FUJISAKI and S.SATO: "Proposal and Evaluation of a New Scheme for Reliable Pitch Extraction of Speech" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 11.14. 473-476 (1990)

[Publications] Hiroya FUJISAKI: "Influence of context and knowkedge on the Perception of Continuous Speech" Proc.International Conference on Spoken Language Processing (ICSLP90),Kobe,Paper 10.9. 417-420 (1990)

[Publications] Katsuhiko SHIRAI: "Speaker Adaptable Phoneme Recognition Selecting Reliable Acoustic Features based on Mutual Information" Proc.of ICSLP90ーInternational conference on Spoken Language Processing. 353-356 (1990)

[Publications] Katsuhiko SHIRAI: "Speech Synthesis Using Superposition of Sinusoidal Waves Generated by Synchronized Oscillators" Proc.of ICSLP90ーInternational conference on Spoken Language Processing. 345-348 (1990)

[Publications] K.GYOUTOKU,(H.KOBATAKE): "Maximum likelihood estimation of speech waveform under nonstationary noise environments" Proc.ICSLP90. 1149-1152 (1990)

[Publications] Hidefumi KOBATAKE: "Enhancement of noisy speech by maximum likelihood estimation" Proc.ICASSP90. (1991)

[Publications] 城風 敏彦、(牧野 正三): "発話全体の連続性を考慮した基本周波数の抽出" 電子情報通信学会論文誌. J73ーA. 1537-1539 (1990)

[Publications] 松尾 広、(牧野 正三): "音素の持続時間モデルに基づく検証法を用いた単語音声認識" 電子情報通信学会論文誌. J73ーDーII. 1936-1944 (1990)

[Publications] 城風 敏彦、(牧野 正三) 城戸 健一: "音声情報の自動獲得機能を持つ分散型大規模音声デ-タベ-ス「KーDB」" 情報処理学会論文誌. 32. 62-70 (1991)

[Publications] 榑松 明: "ATR Japanese Speech Database as a tool of Speech Recognition and Synthesis" SPEECH COMMUNICATION. Volume 9 No.4. 357-357 (1990)

[Publications] 榑松 明: "A Perspective of telephone interpretation research" Proceedign of Pacific Rim International Conference on Artificial Intelligence. 11-16 (1990)

[Publications] 森元 逞(榑松 明): "Spoken language translation toward realizing an automatic telephone interpretation system" Proceeding of InfoーJapan'90(情報処理学会30周年記念国際会議). 553-560 (1990)

板橋秀一筑波大学, 電子・情報工学系, 教授 (70151454)

[Publications] 城風敏彦、(牧野正三): "発話全体の連続性を考慮した基本周波数の抽出" 電子情報通信学会論文誌. J73ーA. 1537-1539 (1990)

[Publications] 松尾広、(牧野正三): "音素の持続時間モデルに基づく検証法を用いた単語音声認識" 電子情報通信学会論文誌. J73ーDーII. 1936-1944 (1990)

[Publications] 城風敏彦、(牧野正三) 城戸健一: "音声情報の自動獲得機能を持つ分散型大規模音声デ-タベ-ス「KーDB」" 情報処理学会論文誌. 32. 62-70 (1991)

[Publications] 榑松明: "ATR Japanese Speech Database as a tool of Speech Recognition and Synthesis" SPEECH COMMUNICATION. Volume 9 No.4. 357-357 (1990)

[Publications] 榑松明: "A Perspective of telephone interpretation research" Proceedign of Pacific Rim International Conference on Artificial Intelligence. 11-16 (1990)

[Publications] 森元逞(榑松明): "Spoken language translation toward realizing an automatic telephone interpretation system" Proceeding of InfoーJapan'90(情報処理学会30周年記念国際会議). 553-560 (1990)