Takato Fujimoto

Tokuda, Nankaku, and Hashimoto Laboratory, Department of Computer Science, Nagoya Institute of Technology

日本語ページ

Profile

Educations

-- present Department of Computer Science, Nagoya Institute of Technology, Japan (PhD)
-- Department of Computer Science, Nagoya Institute of Technology, Japan (Master)
-- Department of Computer Science, Nagoya Institute of Technology, Japan (Bachelor)

Awards

The 7th IEEE Signal Processing Society Tokyo Joint Chapter Student Award
IEEE Nagoya Section Conference Presentation Award 2023
The 2022 IEEE Nagoya Section Excellent student Award
The 22nd Student Presentation Award, Acoustical Society of Japan
The 2019 ASJ-Tokai Meeting student Presentation Award
Overview Lecture Award, the 23rd Tokai region's speech related master's graduation thesis midterm presentation event

Software

-- present SPTK

Publications

International Conference

  1. Autoregressive variational autoencoder with a hidden semi-Markov model-based structured attention for speech synthesis

    2. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7462-7466, Singapore, Singapore, May 2022.

  2. Semi-supervised learning based on hierarchical generative models for end-to-end speech synthesis

    3. Takato Fujimoto, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7644-7648, Barcelona, Spain, May 2020.

  3. Impacts of input linguistic feature representation on Japanese end-to-end speech synthesis

    4. Takato Fujimoto, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of 10th ISCA Speech Synthesis Workshop (SSW10), pp. 166-171, Vienne, Austria, September 2019.

  4. Speech synthesis using WaveNet vocoder based on periodic/aperiodic decomposition

    5. Takato Fujimoto, Takenori Yoshimura, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2018), pp. 644-648, Honolulu, Hawaii, November 2018.

Domestic Conference

  1. Controllable speech synthesis with structured attention based on hidden semi-Markov models

    2. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2024 autumn meeting, pp. 1063-1066, September 2024.

  2. Design of a control interface for text-to-speech synthesis with 55 selectable styles

    3. Sota Nakamura, Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2023 autumn meeting, pp. 1141-1144, September 2023.

  3. Real-time voice convesion in consideration of output delay and time warping transformation

    4. Hikaru Suzuki, Takato Fujimoto, Shinji Takaki, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2023 autumn meeting, pp. 1081-1084, September 2023.

  4. Neural vocoder based on disentangled representation learning to control fundamental frequency

    5. Suzuka Sato, Takato Fujimoto, Yukiya Hono, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2023 autumn meeting, pp. 1061-1064, September 2023.

  5. V2Coder: A neural vocoder based on hierarchical variational autoencoders

    6. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2023 autumn meeting, pp. 1051-1054, September 2023.

  6. Parameter sharing structures in speech synthesis using structured attention based on a hidden semi-Markov model

    7. Ryusei Ishida, Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2022 autumn meeting, pp. 1199-1202, September 2022.

  7. Japanese end-to-end speech synthesis based on hierarchical generative models using semi-supervised learning

    8. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2022 autumn meeting, pp. 1579-1582, September 2022.

  8. Memory reduction methods for sequence-to-sequence speech synthesis using a hidden semi-Markov model based structured attention mechanism

    9. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2022 spring meeting, pp. 969-972, March 2022.

  9. Autoregressive variational autoencoder-based sequence-to-sequence speech synthesis using a hidden semi-Markov model based structured attention mechanism

    10. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2021 autumn meeting, pp. 915-918, September 2021.

  10. Variational autoencoder-based autoregressive sequence-to-sequence speech synthesis considering consistency between training and synthesis

    11. Takato Fujimoto, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2021 spring meeting, pp. 947-950, March 2021.

  11. 出力遅延を考慮したアテンション機構に基づくリアルタイム声質変換

    12. Airi Nishimura, Takato Fujimoto, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Workshop on Informatics 2020, November 2020.

  12. Dirichlet VAE for emotional speech synthesis

    13. Takato Fujimoto, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2020 autumn meeting, pp. 789-790, September 2020.

  13. End-to-end speech synthesis based on hierarchical generative models for semi-supervised learning

    14. Takato Fujimoto, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2020 spring meeting, pp. 1039-1042, March 2020.

  14. A study on singing voice synthesis with attention mechanism using musical score time information

    15. Shumma Murata, Takato Fujimoto, Yukiya Hono, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2019 autumn meeting, pp. 943-944, September 2019.

  15. Impacts of input linguistic features on Japanese end-to-end speech synthesis

    16. Takato Fujimoto, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2019 spring meeting, pp. 1061-1062, March 2019.

  16. Periodic/aperiodic decomposition based speech synthesis using WaveNet vocoder

    17. Takato Fujimoto, Takenori Yoshimura, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda,

    Proceedings of ASJ2018 autumn meeting, pp. 1125-1126, September 2018.

Thesis

  1. End-to-end speech synthesis based on hierarchical generative models for semi-supervised learning

    2. Takato Fujimoto

    Master Thesis, Nagoya Institute of Technology, Feburary 2020.

  2. Periodic/aperiodic decomposed acoustic modeling in speech synthesis based on deep neural networks

    3. Takato Fujimoto

    Graduation Thesis, Nagoya Institute of Technology, February 2018.

Contact

Address

Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya, 466-8555, JAPAN

E-Mail

t.hujimoto.303[at]stn.nitech.ac.jp

Link

Nagoya Institute of Technology
 Tokuda, Nankaku, and Hashimoto Laboratory