Comparison of End-to-End Models for Joint Speaker and Speech Recognition

• Item type: SIG Technical Reports(1)
• 公開日: 2021-02-24
• タイトル: Comparison of End-to-End Models for Joint Speaker and Speech Recognition
• タイトル / 言語: en
• タイトル: Comparison of End-to-End Models for Joint Speaker and Speech Recognition
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: SP1
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: Graduate School of Informatics, Kyoto University / National Institute of Information Communications and Technology (NICT)
• 著者所属: National Institute of Information Communications and Technology (NICT)
• 著者所属: National Institute of Information Communications and Technology (NICT)
• 著者所属: National Institute of Information Communications and Technology (NICT)
• 著者所属: National Institute of Information Communications and Technology (NICT)
• 著者所属(英): en
• 著者所属(英): Graduate School of Informatics, Kyoto University / National Institute of Information Communications and Technology (NICT)
• 著者所属(英): en
• 著者所属(英): National Institute of Information Communications and Technology (NICT)
• 著者所属(英): en
• 著者所属(英): National Institute of Information Communications and Technology (NICT)
• 著者所属(英): en
• 著者所属(英): National Institute of Information Communications and Technology (NICT)
• 著者所属(英): en
• 著者所属(英): National Institute of Information Communications and Technology (NICT)
• 著者名: Kak, Soky; Sheng, Li; Masato, Mimura; Chenhui, Chu; Tatsuya, Kawahara
• 著者名(英): Kak, Soky; Sheng, Li; Masato, Mimura; Chenhui, Chu; Tatsuya, Kawahara
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: In this paper, we investigate the effectiveness of using speaker information on the performance of speaker- imbalanced automatic speech recognition (ASR). We identify major speakers and combine other speakers who have a small size of speech, and make a systematic comparison of three methods that use speaker information for ASR including speaker attribute augmentation (SAug), multi-task learning (MTL), and adversarial learning (AL). We conduct experiments on a large spontaneous speech corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC) and an open Khmer text-to-speech corpus. As a result, we find that the use of speaker clustering information improves ASR performance including new speakers. Moreover, AL achieves better performance and more robustness in the speaker-independent setting compared to the other methods. It reduces errors of the baseline model by 4.32%, 5.46%, and 16.10% for the closed test, open test, and out-of-domain test, respectively.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: In this paper, we investigate the effectiveness of using speaker information on the performance of speaker- imbalanced automatic speech recognition (ASR). We identify major speakers and combine other speakers who have a small size of speech, and make a systematic comparison of three methods that use speaker information for ASR including speaker attribute augmentation (SAug), multi-task learning (MTL), and adversarial learning (AL). We conduct experiments on a large spontaneous speech corpus of the Extraordinary Chambers in the Courts of Cambodia (ECCC) and an open Khmer text-to-speech corpus. As a result, we find that the use of speaker clustering information improves ASR performance including new speakers. Moreover, AL achieves better performance and more robustness in the speaker-independent setting compared to the other methods. It reduces errors of the baseline model by 4.32%, 5.46%, and 16.10% for the closed test, open test, and out-of-domain test, respectively.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10442647
• 書誌情報: 研究報告音声言語情報処理（SLP）; 巻 2021-SLP-136, 号 24, p. 1-5, 発行日 2021-02-24
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8663
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会