* 発表論文 - 2021 [#j9a6ae9c]
#contents
** 論文誌 [#ib3a9611]
+ Yukiya Hono, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``PeriodNet: A Non-Autoregressive Raw Waveform Generative Model With a Structure Separating Periodic and Aperiodic Components,'' IEEE Access, vol. 9, pp. 137599-137612, October, 2021. (Full paper peer reviewed)
&publication(2021/20211005_Journal_IEEE_Yukiya_Hono_paper.pdf, paper);
[[link>https://ieeexplore.ieee.org/document/9559963]]
+ Yukiya Hono, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``Sinsy: A Deep Neural Network-Based Singing Voice Synthesis System,'' IEEE/ACM Transactions on Audio, Speech and Language Processing, vol. 29, pp. 2803-2815, August, 2021. (Full paper peer reviewed)
&publication(2021/20210811_Journal_IEEE_Yukiya_Hono_paper.pdf, paper);
[[link>https://ieeexplore.ieee.org/document/9511835]]
** 国際会議 [#e04bf3ac]
+ %%%Yukiya Hono%%%, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``PeriodNet: A non-autoregressive waveform generation model with a structure separating periodic and aperiodic components,'' 2021 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), pp. 6049-6053, Toronto, Canada, June, 2021. (Full paper peer reviewed)
&publication(2021/20210609_IConference_ICASSP_Yukiya_Hono_paper.pdf, paper);
&publication(2021/20210609_IConference_ICASSP_Yukiya_Hono_poster.pdf, poster);
&publication(2021/20210609_IConference_ICASSP_Yukiya_Hono_slide.pptx, slide);
&publication(2021/20210609_IConference_ICASSP_Yukiya_Hono_video.mp4, video);
[[link (arXiv)>https://arxiv.org/abs/2102.07786]]
(pdfのページ番号がずれてる)
//** 研究会
** 全国大会 [#s719c782]
+ %%%藤本崇人%%%, 橋本佳, 南角吉彦, 徳田恵一,
``隠れセミマルコフモデルによる構造化アテンションを用いた自己回帰型VAEに基づくsequence-to-sequence音声合成,'' 日本音響学会2021年秋季研究発表会, pp. 915-918, 日本, 2021年9月.
//%%%Takato Fujimoto%%%, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, ``Autoregressive variational autoencoder-based sequence-to-sequence speech synthesis using a hidden semi-Markov model based structured attention mechanism,'' Acoustical Society of Japan 2021 Autumn Meeting, pp. 915-918, Japan, September, 2021.
&publication(2021/20210908_DConference_ASJA_Takato_Fujimoto_paper.pdf, paper);
&publication(2021/20210908_DConference_ASJA_Takato_Fujimoto_abst.pdf, abst);
&publication(2021/20210908_DConference_ASJA_Takato_Fujimoto_slide.pptx, slide);
(オンライン開催)
+ %%%法野行哉%%%, 加藤大誠, 橋本佳, 大浦圭一郎,南角吉彦, 徳田恵一
``発声タイミングのずれを考慮したsequence-to-sequence歌声合成,'' 日本音響学会2021年秋季研究発表会, pp. 911-914, 日本, 2021年9月.
//%%%Yukiya Hono%%%, Taisei Kato, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``Sequence-to-sequence singing voice synthesis considering vocal timing fluctuation,'' Acoustical Society of Japan 2021 Autumn Meeting, pp. 911-914, Japan, September, 2021.
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_paper-2.pdf, paper);
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_abst-2.pdf, abst);
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_slide-2.pptx, slide);
(オンライン開催)
+ %%%法野行哉%%%, 橋本佳, 大浦圭一郎,南角吉彦, 徳田恵一
``DNN歌声合成のための調子はずれ補正,'' 日本音響学会2021年秋季研究発表会, pp. 907-910, 日本, 2021年9月.
//%%%Yukiya Hono%%%, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``Automatic pitch correction of tuneless singing for DNN-based singing voice synthesis,'' Acoustical Society of Japan 2021 Autumn Meeting, pp. 907-910, Japan, September, 2021.
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_paper-1.pdf, paper);
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_abst-1.pdf, abst);
&publication(2021/20210908_DConference_ASJA_Yukiya_Hono_slide-1.pptx, slide);
(オンライン開催)
+ %%%高木信二%%%, 牛田光一, 橋本佳, 南角吉彦, 徳田恵一,
``因子分析に基づくHSMMを利用した構造化アテンション音声合成,'' 日本音響学会2021年秋季研究発表会, pp. 871-874, 日本, 2021年9月.
//%%%Shinji Takaki%%%, Koichi Ushida, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, ``A factor analyzed voice model for sequence-to-sequence speech synthesis using a hidden semi-Markov model based structured attention mechanism,'' Acoustical Society of Japan 2021 Autumn Meeting, pp. 871-874, Japan, September, 2021.
&publication(2021/20210907_DConference_ASJA_Shinji_Takaki_paper.pdf, paper);
&publication(2021/20210907_DConference_ASJA_Shinji_Takaki_abst.pdf, abst);
&publication(2021/20210907_DConference_ASJA_Shinji_Takaki_slide.pptx, slide);
(オンライン開催)
+ %%%藤本崇人%%%, 橋本佳, 南角吉彦, 徳田恵一,
``学習時と合成時の一貫性を考慮したVAEに基づく自己回帰型sequence-to-sequence音声合成,'' 日本音響学会2021年春季研究発表会, pp. 947-950, 日本, 2021年3月.
//%%%Takato Fujimoto%%%, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, ``Variational autoencoder-based autoregressive sequence-to-sequence speech synthesis considering consistency between training and synthesis,'' Acoustical Society of Japan 2021 Spring Meeting, pp. 947-950, Japan, March, 2021.
&publication(2021/20210312_DConference_ASJS_Takato_Fujimoto_paper.pdf, paper);
&publication(2021/20210312_DConference_ASJS_Takato_Fujimoto_abst.pdf, abst);
&publication(2021/20210312_DConference_ASJS_Takato_Fujimoto_slide.pptx, slide);
(オンライン開催)
+ %%%角谷健太%%%, 吉村建慶, 高木信二, 橋本佳, 大浦圭一郎, 南角吉彦, 徳田恵一,
``隠れセミマルコフモデルに基づく構造化アテンションを用いたSequence-to-Sequence音声合成,'' 日本音響学会2021年春季研究発表会, pp. 943-946, 日本, 2021年3月.
//%%%Takato Fujimoto%%%, Yoshimura Takenori, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``Sequence-to-sequence speech synthesis using a hidden semi-Markov model based structured attention mechanism,'' Acoustical Society of Japan 2021 Spring Meeting, pp. 943-946, Japan, March, 2021.
&publication(2021/20210312_DConference_ASJS_Kenta_Sumiya_paper.pdf, paper);
&publication(2021/20210312_DConference_ASJS_Kenta_Sumiya_abst.pdf, abst);
&publication(2021/20210312_DConference_ASJS_Kenta_Sumiya_slide.pptx, slide);
(オンライン開催)
+ %%%法野行哉%%%, 高木信二, 橋本佳, 大浦圭一郎, 南角吉彦, 徳田恵一,
``周期・非周期成分の分離に基づくニューラルボコーダによる音声波形のモデル化の検討,'' 日本音響学会2021年春季研究発表会, pp. 861-864, 日本, 2021年3月.
//%%%Yukiya Hono%%%, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``An investigation of modeling speech waveform by neural vocoder based on periodic/aperiodic decomposition,'' Acoustical Society of Japan 2021 Spring Meeting, pp. 861-864, Japan, March, 2021.
&publication(2021/20210311_DConference_ASJS_Yukiya_Hono_paper.pdf, paper);
&publication(2021/20210311_DConference_ASJS_Yukiya_Hono_abst.pdf, abst);
&publication(2021/20210311_DConference_ASJS_Yukiya_Hono_slide.pptx, slide);
(オンライン開催)
+ %%%岩田康平%%%, 高木信二, 橋本佳, 南角吉彦, 徳田恵一,
``勾配ブースティング決定木を用いた音声合成手法の検討,'' 日本音響学会2021年春季研究発表会, pp. 813-814, 日本, 2021年3月.
//%%%Kohei Iwata%%%, Shinji Takaki, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, ``An investigation of speech synthesis based on gradient boosting decision trees,'' Acoustical Society of Japan 2021 Spring Meeting, pp. xx-xx, Japan, March, 2021.
&publication(2021/20210310_DConference_ASJS_Kohei_Iwata_paper.pdf, paper);
&publication(2021/20210310_DConference_ASJS_Kohei_Iwata_abst.pdf, abst);
&publication(2021/20210310_DConference_ASJS_Kohei_Iwata_slide.pptx, slide);
(オンライン開催)
** 学位論文 [#sfb5bb22]
+ %%%白木祐弥%%%,
``デコーディング時を想定した系列識別学習によるEnd-to-End音声認識''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``END-TO-END SPEECH RECOGNITION WITH SEQUENCE DISCRIMINATIVE TRAINING UNDER CONSTRAINTS OF DECODING ALGORITHMS,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Yuya_Shiraki_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Yuya_Shiraki_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Yuya_Shiraki_abst.pdf, abst);
+ %%%伊良部瑞基%%%,
``汎用言語表現モデルを用いたニューラルネットワークに基づく音声合成''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``TEXT-TO-SPEECH SYNTHESIS BASED ON NEURAL NETWORKS USING GENERAL PURPOSE LANGUAGE MODELS,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Mizuki_Irabu_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Mizuki_Irabu_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Mizuki_Irabu_abst.pdf, abst);
+ %%%伊藤哲平%%%,
``リアルタイム声質変換における聴覚フィードバックとユーザビリティに関する調査''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``INVESTIGATING RELATION BETWEEN AUDITORY FEEDBACK AND USABILITY IN REAL-TIME VOICE CONVERSION,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Teppei_Ito_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Teppei_Ito_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Teppei_Ito_abst.pdf, abst);
+ %%%加藤あまね%%%,
``Dirichlet分布を用いたVariational Auto-Encoderによる音声合成のための話者適応'',
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SPEAKER ADAPTATION FOR SPEECH SYNTHESIS BASED ON VARIATIONAL AUTO-ENCODERS USING DIRICHLET DISTRIBUTIONS,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Amane_Kato_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Amane_Kato_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Amane_Kato_abst.pdf, abst);
+ %%%山田淳司%%%,
``音素事後確率を用いたノンパラレル声質変換における基本周波数の利用''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``USE OF FUNDAMENTAL FREQUENCIES IN NON-PARALLEL VOICE CONVERSION BASED ON POSTERIOR PROBABILITIES,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Atsushi_Yamada_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Atsushi_Yamada_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Atsushi_Yamada_abst.pdf, abst);
+ %%%東浩幸%%%,
``話者認識におけるNoisy Studentを用いたモデル学習法の検討''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SPEAKER RECOGNITION BASED ON MODEL TRAINING WITH NOISY STUDENT,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Hiroyuki_Higashi_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Hiroyuki_Higashi_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Hiroyuki_Higashi_abst.pdf, abst);
+ %%%八野田響%%%,
``音量の正規化構造を導入したニューラルネットワークに基づく音声合成''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SPEECH SYNTHESIS BASED ON NEURAL NETWORKS INTEGRARING LOUDNESS NORMALIZATION,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Hibiki_Hachinoda_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Hibiki_Hachinoda_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Hibiki_Hachinoda_abst.pdf, abst);
+ %%%野竹浩二朗%%%,
多様な発話スタイルを表現するための敵対的生成モデルに基づく音声合成''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SPEECH SYNTHESIS TO EXPRESS VARIOUS SPEAKING STYLES BASED ON GENERATIVE ADVERSARIAL NETWORKS,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Kojiro_Notake_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Kojiro_Notake_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Kojiro_Notake_abst.pdf, abst);
+ %%%加藤大誠%%%,
``発声タイミングのずれを考慮したアテンション機構に基づく歌声合成の検討''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SINGING VOICE SYNTHESIS BASED ON ATTENTION MECHANISMS CONSIDERING VOCAL TIMING FLUCTUATION,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Taisei_Kato_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Taisei_Kato_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Taisei_Kato_abst.pdf, abst);
+ %%%石田シゲオ%%%,
``音声入力におけるプライバシー保護のためのサウンドマスキング''
卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``SOUND MASKING FOR PRIVACY PROTECTION IN SPEECH INPUT,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Shigeo_Ishida_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Shigeo_Ishida_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Shigeo_Ishida_abst.pdf, abst);
+ %%%砂田憲廣%%%, ``高サンプリングレート時におけるWaveNetボコーダの最適モデル構造の検討'' 卒業論文, 名古屋工業大学, 2021年2月.
//%%% %%%, ``OPTIMIZING MODEL STRUCTURES OF WAVENET VOCODERS AT A HIGH SAMPLING RATE,'' February, 2021.
&publication(2021/20210215_Thesis_Bachelor_Norihiro_Sunada_paper.pdf, paper);
&publication(2021/20210215_Thesis_Bachelor_Norihiro_Sunada_slide.pptx, slide);
&publication(2021/20210215_Thesis_Bachelor_Norihiro_Sunada_abst.pdf, abst);
+ %%%佐藤優介%%%,
``ニューラルネットワークに基づく音声合成における言語依存層と話者コードを用いたクロスリンガル話者適応,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Yusuke Sato%%%, ``CROSS-LINGUAL SPEAKER ADAPTATION IN SPEECH SYNTHESIS BASED ON NEURAL NETWORKS USING LANGUAGE-DEPENDENT LAYERS AND SPEAKER CODES,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Yusuke_Sato_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Yusuke_Sato_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Yusuke_Sato_abst.pdf, abst);
+ %%%和田蒼汰%%%,
``周期・非周期信号を潜在変数とするDeep AutoEncoderに基づくニューラルボコーダ,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Sota Wada%%%, ``ACOUSTIC MODEL STRUCTURES FOR REALTIME INSTRUMENT-PLAYING SOUND SYNTHESIS,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Sota_Wada_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Sota_Wada_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Sota_Wada_abst.pdf, abst);
+ %%%中村洋太%%%,
``リアルタイム楽器演奏音合成のための音響モデル構成法,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Yota Nakamura%%%, ``ACOUSTIC MODEL STRUCTURES FOR REALTIME INSTRUMENT-PLAYING SOUND SYNTHESIS,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Yota_Nakamura_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Yota_Nakamura_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Yota_Nakamura_abst.pdf, abst);
+ %%%張甘辰星%%%,
``Dirichlet分布を用いたVariational Auto-Encoderに基づく感情音声合成システム,'' 修士論文, 名古屋工業大学, 2021年2月.
//%%%Zhang Ganchenxing%%%, ``EMOTIONAL SPEECH SYNTHESIS SYSTEM BASED ON DIRICHLET VARIATIONAL AUTO-ENCODER,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Zhang_Ganchenxing_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Zhang_Ganchenxing_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Zhang_Ganchenxing_abst.pdf, abst);
+ %%%牛田光一%%%,
``因子分析の構造を用いたニューラルネットワークに基づく発話スタイルを考慮した音声合成,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Koichi Ushida%%%, ``SPEECH SYNTHESIS CONSIDERING SPEAKING STYLES BASED ON NEURAL NETWORKS USING FACTOR ANALYSIS STRUCTURES,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Koichi_Ushida_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Koichi_Ushida_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Koichi_Ushida_abst.pdf, abst);
+ %%%角谷健太%%%,
``End-to-End音声合成のための隠れセミマルコフモデルに基づく構造化アテンションを用いた音響モデリング,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Kenta Sumiya%%%, ``ACOUSTIC MODELING USING A HIDDEN SEMI-MARKOV MODEL BASED STRUCTURED ATTENTION MECHANISM FOR END-TO-END SPEECH SYNTHESIS,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Kenta_Sumiya_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Kenta_Sumiya_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Kenta_Sumiya_abst.pdf, abst);
+ %%%清水達也%%%,
``話者認識のための系列変分オートエンコーダに基づく純粋な話者特徴の抽出,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Tatsuya Shimizu%%%, ``PURE SPEAKER FEATURE REPRESENTATION BASED ON SEQUENTIAL VARIATIONAL AUTOENCODER FOR SPEAKER RECOGNITION,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Tatsuya_Shimizu_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Tatsuya_Shimizu_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Tatsuya_Shimizu_abst.pdf, abst);
+ %%%木下耕介%%%,
``幾何学的変動に頑健な画像認識のためのAttention機構を内包した深層生成モデル,''
修士論文, 名古屋工業大学, 2021年2月.
//%%%Kousuke Kinoshita%%%, ``DEEP GENERATIVE MODELS WITH ATTENTION MECHANISM FOR ROBUST IMAGE RECOGNITION AGAINST GEOMETRIC VARIATIONS,'' Master thesis, Nagoya Institute of Technology, February, 2021.
&publication(2021/20210205_Thesis_Master_Kosuke_Kinoshita_paper.pdf, paper);
&publication(2021/20210205_Thesis_Master_Kosuke_Kinoshita_slide.pptx, slide);
&publication(2021/20210205_Thesis_Master_Kosuke_Kinoshita_abst.pdf, abst);
** プレプリント [#mcdf6a12]
+ %%%Yukiya Hono%%%, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, ``PeriodNet: A non-autoregressive waveform generation model with a structure separating periodic and aperiodic components,'' arXiv preprint arXiv:2102.07786, Feburary, 2021.
&publication(2021/20210215_Preprint_arXiv_Yukiya_Hono_paper.pdf, paper);
[[link>https://arxiv.org/abs/2102.07786]]
** 著書 [#g0d1a79d]
+ 南角吉彦（分担執筆）, ``音響学講座6 音声（上）,'' 日本音響学会編, コロナ社, 2021年9月22日.
**過去の発表論文 [#ifa8da9a]
#ls2(ホーム/発表論文/,reverse);
