WEKO3
アイテム
{"_buckets": {"deposit": "583c9d08-2e3d-4950-8b39-d8ad55f4b457"}, "_deposit": {"created_by": 4, "id": "4607", "owners": [4], "pid": {"revision_id": 0, "type": "depid", "value": "4607"}, "status": "published"}, "_oai": {"id": "oai:naist.repo.nii.ac.jp:00004607", "sets": ["36"]}, "author_link": ["11885", "11886", "11887", "11888", "11889", "11890"], "item_1698715929687": {"attribute_name": "会議情報", "attribute_value_mlt": [{"subitem_conference_country": "USA", "subitem_conference_date": {"subitem_conference_date_language": "en", "subitem_conference_period": "May 7-11, 2001"}, "subitem_conference_names": [{"subitem_conference_name": "ICASSP2001: IEEE International Conference on Acoustics, Speech and Signal Processing", "subitem_conference_name_language": "en"}], "subitem_conference_places": [{"subitem_conference_place": "Salt Lake City Utah", "subitem_conference_place_language": "en"}]}]}, "item_9_biblio_info_7": {"attribute_name": "書誌情報", "attribute_value_mlt": [{"bibliographicIssueDates": {"bibliographicIssueDate": "2001-05", "bibliographicIssueDateType": "Issued"}, "bibliographicPageEnd": "344", "bibliographicPageStart": "341"}]}, "item_9_description_5": {"attribute_name": "抄録", "attribute_value_mlt": [{"subitem_description": "Describes an efficient method for unsupervised speaker adaptation. This method is based on (1) selecting a subset of speakers who are acoustically close to a test speaker, and (2) calculating adapted model parameters according to the previously stored sufficient HMM statistics of the selected speakers\u0027 data. In this method, only a few unsupervised test speaker\u0027s data are required for the adaptation. Also, by using the sufficient HMM statistics of the selected speakers\u0027 data, a quick adaptation can be done. Compared with a pre-clustering method, the proposed method can obtain a more optimal speaker cluster because the clustering result is determined according to test speaker\u0027s data on-line. Experimental results show that the proposed method attains better improvement than MLLR from the speaker independent model. Moreover the proposed method utilizes only one unsupervised sentence utterance, while MLLR usually utilizes more than ten supervised sentence utterances", "subitem_description_language": "en", "subitem_description_type": "Abstract"}]}, "item_9_publisher_8": {"attribute_name": "出版者", "attribute_value_mlt": [{"subitem_publisher": "IEEE", "subitem_publisher_language": "en"}]}, "item_9_relation_11": {"attribute_name": "ISBN", "attribute_value_mlt": [{"subitem_relation_type_id": {"subitem_relation_type_id_text": "0780370414", "subitem_relation_type_select": "ISBN"}}]}, "item_9_relation_13": {"attribute_name": "DOI", "attribute_value_mlt": [{"subitem_relation_type": "isIdenticalTo", "subitem_relation_type_id": {"subitem_relation_type_id_text": "https://doi.org/10.1109/ICASSP.2001.940837", "subitem_relation_type_select": "DOI"}}]}, "item_9_rights_14": {"attribute_name": "権利", "attribute_value_mlt": [{"subitem_rights": "Copyright 2001 IEEE", "subitem_rights_language": "en"}]}, "item_9_source_id_10": {"attribute_name": "ISSN", "attribute_value_mlt": [{"subitem_source_identifier": "1520-6149", "subitem_source_identifier_type": "ISSN"}]}, "item_9_version_type_16": {"attribute_name": "著者版フラグ", "attribute_value_mlt": [{"subitem_version_resource": "http://purl.org/coar/version/c_970fb48d4fbd8a85", "subitem_version_type": "VoR"}]}, "item_access_right": {"attribute_name": "アクセス権", "attribute_value_mlt": [{"subitem_access_right_uri": "open access"}]}, "item_creator": {"attribute_name": "著者", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "Yoshizawa, Shinichi", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11885", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Baba, Akira", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11886", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Matsunami, Kanako", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11887", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Mera, Yuichiro", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11888", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Yamada, Miichi", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11889", "nameIdentifierScheme": "WEKO"}]}, {"creatorNames": [{"creatorName": "Shikano, Kiyohiro", "creatorNameLang": "en"}], "nameIdentifiers": [{"nameIdentifier": "11890", "nameIdentifierScheme": "WEKO"}]}]}, "item_files": {"attribute_name": "ファイル情報", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_date", "date": [{"dateType": "Available", "dateValue": "2023-03-02"}], "displaytype": "detail", "download_preview_message": "", "file_order": 0, "filename": "ICASSP_2001_341.pdf", "filesize": [{"value": "471.8 kB"}], "format": "application/pdf", "future_date_message": "", "is_thumbnail": false, "licensetype": "license_note", "mimetype": "application/pdf", "size": 471800.0, "url": {"label": "fulltext", "objectType": "fulltext", "url": "https://naist.repo.nii.ac.jp/record/4607/files/ICASSP_2001_341.pdf"}, "version_id": "5eba15cb-645e-490c-ac19-f40861ca10fa"}]}, "item_language": {"attribute_name": "言語", "attribute_value_mlt": [{"subitem_language": "eng"}]}, "item_resource_type": {"attribute_name": "資源タイプ", "attribute_value_mlt": [{"resourcetype": "conference paper", "resourceuri": "http://purl.org/coar/resource_type/c_5794"}]}, "item_title": "Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers", "item_titles": {"attribute_name": "タイトル", "attribute_value_mlt": [{"subitem_title": "Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers", "subitem_title_language": "en"}]}, "item_type_id": "9", "owner": "4", "path": ["36"], "permalink_uri": "http://hdl.handle.net/10061/8039", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2012-08-22"}, "publish_date": "2012-08-22", "publish_status": "0", "recid": "4607", "relation": {}, "relation_version_is_last": true, "title": ["Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers"], "weko_shared_id": -1}
Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers
http://hdl.handle.net/10061/8039
http://hdl.handle.net/10061/8039ec18bbe4-30b4-4628-af69-5cd91573bf68
名前 / ファイル | ライセンス | アクション |
---|---|---|
fulltext (471.8 kB)
|
|
Item type | 会議発表論文 / Conference Paper(1) | |||||
---|---|---|---|---|---|---|
公開日 | 2012-08-22 | |||||
タイトル | ||||||
タイトル | Unsupervised Speaker Adaptation Based on Sufficient HMM Statistics of Selected Speakers | |||||
言語 | ||||||
言語 | eng | |||||
資源タイプ | ||||||
資源タイプ | conference paper | |||||
アクセス権 | ||||||
著者 |
Yoshizawa, Shinichi
× Yoshizawa, Shinichi× Baba, Akira× Matsunami, Kanako× Mera, Yuichiro× Yamada, Miichi× Shikano, Kiyohiro |
|||||
抄録 | ||||||
内容記述タイプ | Abstract | |||||
内容記述 | Describes an efficient method for unsupervised speaker adaptation. This method is based on (1) selecting a subset of speakers who are acoustically close to a test speaker, and (2) calculating adapted model parameters according to the previously stored sufficient HMM statistics of the selected speakers' data. In this method, only a few unsupervised test speaker's data are required for the adaptation. Also, by using the sufficient HMM statistics of the selected speakers' data, a quick adaptation can be done. Compared with a pre-clustering method, the proposed method can obtain a more optimal speaker cluster because the clustering result is determined according to test speaker's data on-line. Experimental results show that the proposed method attains better improvement than MLLR from the speaker independent model. Moreover the proposed method utilizes only one unsupervised sentence utterance, while MLLR usually utilizes more than ten supervised sentence utterances | |||||
書誌情報 |
p. 341-344, 発行日 2001-05 |
|||||
会議情報 | ||||||
会議名 | ICASSP2001: IEEE International Conference on Acoustics, Speech and Signal Processing | |||||
開催期間 | May 7-11, 2001 | |||||
開催地 | Salt Lake City Utah | |||||
開催国 | USA | |||||
出版者 | ||||||
出版者 | IEEE | |||||
ISSN | ||||||
収録物識別子タイプ | ISSN | |||||
収録物識別子 | 1520-6149 | |||||
ISBN | ||||||
識別子タイプ | ISBN | |||||
関連識別子 | 0780370414 | |||||
DOI | ||||||
関連タイプ | isIdenticalTo | |||||
識別子タイプ | DOI | |||||
関連識別子 | https://doi.org/10.1109/ICASSP.2001.940837 | |||||
権利 | ||||||
権利情報 | Copyright 2001 IEEE | |||||
著者版フラグ | ||||||
出版タイプ | VoR |