Using Continuous Representation of Various Linguistic Units for Recurrent Neural Network based TTS Synthesis

• Item type: SIG Technical Reports(1)
• 公開日: 2016-01-29
• タイトル: Using Continuous Representation of Various Linguistic Units for Recurrent Neural Network based TTS Synthesis
• タイトル / 言語: en
• タイトル: Using Continuous Representation of Various Linguistic Units for Recurrent Neural Network based TTS Synthesis
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: 一般：対雑音、合成
• 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: National Institute of Informatics
• 著者所属: National Institute of Informatics
• 著者所属: National Institute of Informatics
• 著者所属(英): en
• 著者所属(英): National Institute of Informatics
• 著者所属(英): en
• 著者所属(英): National Institute of Informatics
• 著者所属(英): en
• 著者所属(英): National Institute of Informatics
• 著者名: Xin, Wang; Shinji, Takaki; Junichi, Yamagishi
• 著者名(英): Xin, Wang; Shinji, Takaki; Junichi, Yamagishi
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Building high-quality text-to-speech (TTS) systems without expert knowledge of the target language and/or manual time-consuming annotation of speech and text data is an important and challenging research topic in speech synthesis. Recently, the distributed representation of raw word inputs, called “word embedding”, have been used in various natural language processing tasks with success. Moreover, the word-embedding vectors have recently been used as the additional or alternative linguistic input features to a neural-network-based acoustic model for TTS systems. Since word-embedding approaches may provide means for obtaining effective linguistic representations from texts without requiring specialized knowledge of the language and/or manual time-consuming annotation, we further investigated the use of the word-embedding for neural-network-based TTS systems in two new directions. First, in addition to the standard word embedding vectors, we attempted to use phoneme, syllable, and phrase embedding vectors to verify whether continuous representations of these linguistic units may improve the segmental and suprasegmental quality of synthetic speech. Second, we examined the impact of normalization methods on the obtained embedded vectors before they were feed into the neural-network-based acoustic model.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: Building high-quality text-to-speech (TTS) systems without expert knowledge of the target language and/or manual time-consuming annotation of speech and text data is an important and challenging research topic in speech synthesis. Recently, the distributed representation of raw word inputs, called “word embedding”, have been used in various natural language processing tasks with success. Moreover, the word-embedding vectors have recently been used as the additional or alternative linguistic input features to a neural-network-based acoustic model for TTS systems. Since word-embedding approaches may provide means for obtaining effective linguistic representations from texts without requiring specialized knowledge of the language and/or manual time-consuming annotation, we further investigated the use of the word-embedding for neural-network-based TTS systems in two new directions. First, in addition to the standard word embedding vectors, we attempted to use phoneme, syllable, and phrase embedding vectors to verify whether continuous representations of these linguistic units may improve the segmental and suprasegmental quality of synthetic speech. Second, we examined the impact of normalization methods on the obtained embedded vectors before they were feed into the neural-network-based acoustic model.
• 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 2016-SLP-110, 号 7, p. 1-6, 発行日 2016-01-29
• 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