Kei Sawada

Profile

Nagoya Institute of Technology
Tokuda and Nankaku Laboratory
Kei Sawada

Professional Experience

Education

Project

Award

Society

Contact

Publication

Journal Paper

  1. Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "A Bayesian framework for image recognition based on hidden Markov eigen-image models," IEEJ Transactions on Electrical and Electronic Engineering, Vol.13, Issue 9, pp.1335-1347, DOI:10.1002/tee.22700, 2018.09 [paper] [link]
  2. Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Constructing text-to-speech systems for languages with unknown pronunciations," Acoustical Science and Technology, Vol.39, Issue 2, pp.119-129, DOI: 10.1250/ast.39.119, 2018.03 [paper] [link]
  3. Kei Sawada, Akira Tamamori, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "A Bayesian approach to image recognition based on separable lattice hidden Markov models," IEICE Transactions on Information and Systems, Vol.E99-D, No.12, pp.3119-3131, DOI: 10.1587/transinf.2016EDP7112, 2016.12 [paper] [link]

International Conference

  1. Koki Senda, Yukiya Hono, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Singing voice conversion using posted waveform data on music social media," Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2018), Hawaii, USA, 2018.11
  2. Takayuki Kasugai, Yoshinari Tsuzuki, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on convolutional neural networks using features generated from separable lattice hidden Markov models," Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2018), Hawaii, USA, 2018.11
  3. Kei Sawada, Takenori Yoshimura, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "The NITech text-to-speech system for the Blizzard Challenge 2018," Blizzard Challenge 2018 Workshop, Hyderabad, India, 2018.09 [paper] [slide] [speech samples] [link]
  4. Eiji Ichikawa, Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on separable lattice HMMs using a deep neural network for output probability distributions," 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2018), pp.3021-3025, Calgary, Canada, 2018.04
  5. Kei Sawada, Keiichi Tokuda, Simon King, and Alan W Black, "The Blizzard Machine Learning Challenge 2017," 2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU 2017), pp.331-337, Okinawa, Japan, 2017.12 [paper] [slide] [link] [copyright]
  6. Kei Sawada, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "The NITech text-to-speech system for the Blizzard Challenge 2017," Blizzard Challenge 2017 Workshop, Stockholm, Sweden, 2017.08 [paper] [slide] [speech samples] [link]
  7. Yoshinari Tsuzuki, Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on discriminative models using features generated from separable lattice HMMs," 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2017), pp.2607-2611, New Orleans, USA, 2017.03 [link]
  8. Kei Sawada, Chiaki Asai, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "The NITech text-to-speech system for the Blizzard Challenge 2016," Blizzard Challenge 2016 Workshop, Cupertino, USA, 2016.09 [paper] [slide] [link]
  9. Kei Sawada, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "The NITECH HMM-based text-to-speech system for the Blizzard Challenge 2015," Blizzard Challenge 2015 Workshop, Berlin, Germany, 2015.09 [paper] [slide] [link]
  10. Kei Sawada, Shinji Takaki, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "Overview of NITECH HMM-based text-to-speech system for Blizzard Challenge 2014," Blizzard Challenge 2014 Workshop, Singapore, Singapore, 2014.09 [paper] [slide] [link]
  11. Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on hidden Markov eigen-image models using variational Bayesian method," Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC 2013), pp.1-8, Kaohsiung, Taiwan, 2013.10 [paper] [slide] [link]
  12. Shinji Takaki, Kei Sawada, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "Overview of NITECH HMM-based speech synthesis system for Blizzard Challenge 2013," Blizzard Challenge 2013 Workshop, Barcelona, Spain, 2013.09 [paper] [link]
  13. Shinji Takaki, Kei Sawada, Kei Hashimoto, Keiichiro Oura, and Keiichi Tokuda, "Overview of NIT HMM-based speech synthesis system for Blizzard Challenge 2012," Blizzard Challenge 2012 Workshop, Portland, USA, 2012.09 [paper] [link]
  14. Kei Sawada, Akira Tamamori, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Face recognition based on separable lattice 2-D HMMs using variational bayesian method," 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2012), pp.2205-2208, Kyoto, Japan, 2012.03 [paper] [poster] [link] [copyright]

Technical Report

  1. Chiaki Asai, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Designing linguistic features for expressive speech synthesis using audiobooks," IEICE Technical Report, vol.116, no.414, SP2016-70, pp.35-40, 2017.01 [link]
  2. Yoshinari Tsuzuki, Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on discriminative models using features generated from separable lattice HMMs," IEICE Technical Report, vol.116, no.89, PRMU2016-36, pp.7-12, 2016.06 [link]
  3. Masato Sukegawa, Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Parameter sharing structures of separable lattice HMMs using mixture output distributions for image recognition," IEICE Technical Report, vol.115, no.456, PRMU2015-138, pp.37-42, 2016.02 [link]
  4. Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Evaluation of text-to-speech system construction for unknown-pronunciation languages," IEICE Technical Report, vol.115, no.346, SP2015-80, pp.93-98, 2015.12 [slide] [speech samples] [movie] [link]
  5. Kei Sawada, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Image recognition based on hidden Markov eigen-image models with the variational Bayesian method," IEICE Technical Report, vol.112, no.441, PRMU2012-165, pp.155-160, 2013.02 [slide] [link]
  6. Kei Sawada, Akira Tamamori, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "Face recognition based on separable lattice 2-D HMMs with variational Bayesian method," IEICE Technical Report, vol.111, no.317, PRMU2011-120, pp.125-130, 2011.11 [slide] [link]

Domestic Conference

  1. Kei Sawada, Takenori Yoshimura, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Overview of the NITech text-to-speech system for the Blizzard Challenge 2018," Acoustical Society of Japan 2018 Autumn Meeting, pp.1091-1094, 2018.09 [paper] [slide] [speech samples] [link]
  2. Kei Sawada, Keiichi Tokuda, Simon King, and Alan W Black, "Overview of the Blizzard Machine Learning Challenge 2017," Acoustical Society of Japan 2018 Spring Meeting, pp.1509-1510, 2018.03 [paper] [slide] [link]
  3. Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Overview of the NITech text-to-speech system for the Blizzard Challenge 2017," Acoustical Society of Japan 2017 Autumn Meeting, pp.287-290, 2017.09 [paper] [poster] [speech samples] [link]
  4. Yukiya Hono, Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, Keiichi Tokuda, Daisuke Kondo, and Daisuke Ishikawa, "Singing voice conversion using post data in music SNS," Acoustical Society of Japan 2017 Autumn Meeting, pp.209-210, 2017.09 [link]
  5. Kei Sawada, Kazuki Igami, Chiaki Asai, Yusuke Sato, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Automatic construction of training corpus using audiobooks for statistical parametric speech synthesis," Acoustical Society of Japan 2016 Spring Meeting, pp.219-220, 2016.03 (The 13th Student Presentation Award) [paper] [slide] [link]
  6. Kei Sawada, Kei Hashimoto, Keiichiro Oura, Yoshihiko Nankaku, and Keiichi Tokuda, "Investigation of text-to-speech system construction in unknown-pronunciation language," Acoustical Society of Japan 2015 Autumn Meeting, pp.231-232, 2015.09 [paper] [slide] [link]
  7. Kei Sawada, Akira Tamamori, Kei Hashimoto, Yoshihiko Nankaku, and Keiichi Tokuda, "A training algorithm based on variational Bayesian method using deterministic annealing process for separable lattice 2-D HMMs," The 74th National Convention of IPSJ, vol.2, pp.409-410, 2012.03 [paper] [slide] [link] [copyright]

Thesis

  1. Kei Sawada, "A statistical approach to speech synthesis and image recognition based on hidden Markov models," Doctor Thesis, Nagoya Institute of Technology, 2018.02
  2. Kei Sawada, "A Bayesian approach to image recognition based on hidden Markov eigen-image models," Master Thesis, Nagoya Institute of Technology, 2013.02
  3. Kei Sawada, "Face recognition based on separable lattice 2-D HMMs with variational Bayesian method," Graduation Thesis, Nagoya Institute of Technology, 2011.02

Exhibition

  1. MMDAgent DAY!, 2016.10 [link]
  2. Science Agora 2012, 2012.11 [link]

Link