About

Hi, I am a research scientist at Google DeepMind, working on multimodal language modeling in the Speech Team. My research interests include audio generation, large language models and multilinguality. I obtained my PhD at the Graduate school of Information Science and Technology, the University of Tokyo, Japan.

Email (office): tsaeki [at] google.com
Email (personal): saefrospace [at] gmail.com

Education

• Apr. 2021 - Mar. 2024
  Ph.D. Degree in Information Science and Technology, the University of Tokyo, Japan
  Department of Creative Informatics, Graduate School of Information Science and Technology
  Supervisor: Prof. Hiroshi Saruwatari
• Apr. 2019 - Mar. 2021
  Master’s Degree in Information Science and Technology, the University of Tokyo, Japan
  Department of Creative Informatics, Graduate School of Information Science and Technology
  Supervisor: Prof. Hiroshi Saruwatari
• Apr. 2015 - Mar. 2019
  Bachelor’s Degree in Engineering, the University of Tokyo, Japan
  Department of Aeronautics and Astronautics, Faculty of Engineering

Awards & Honours

• June 2024
  IEICE ISS Young Researcher’s Award in Speech Field [link]
  IEICE Speech Committee
• Mar. 2024
  Dean’s Award [link]
  Graduate School of Information Science and Technology, The University of Tokyo
• July 2022
  Yamashita SIG Research Award [link]
  Information Processing Society of Japan (IPSJ)
• June 2022
  Best Paper Award from IEICE [link]
  The Institute of Electronics, Information and Communication Engineers (IEICE)
• Mar. 2022
  Ranked 1st Place in 10/16 Metrics [link]
  VoiceMOS Challenge 2022 at INTERSPEECH 2022
• Mar. 2022
  Telecom System Technology Award for Students [link]
  The Telecommunication Advancement Foundation
• Oct. 2021
  Best Student Presentation Award [link]
  Acoustical Society of Japan (ASJ)
• Mar. 2021
  Best Student Poster Award [link]
  IEICE Speech Committee
• Feb. 2019
  Award for Excellence (2nd Place)
  Recruit Holdings NLP Hackathon

Grants & Scholarships

• Aug. 2022
  Google East Asia Student Travel Grants
  Google
• Mar. 2022
  UTokyo-TOYOTA Study Abroad Scholarship in AI Field
  The University of Tokyo
• Apr. 2022 - Mar. 2024
  Research Fellowship for Young Scientists (DC2)
  Japan Society for the Promotion of Science (JSPS)
• July 2021
  Tobitate Study Abroad Initiative (Declined due to COVID19)
  Ministry of Education, Culture, Sports, Science and Technology
• June 2021 - Mar. 2022
  TOYOTA/Dwango AI Scholarship
  The University of Tokyo

Experience

• Nov. 2024 - present
  Google DeepMind, New York, Research Scientist
  Researching on multimodal large language models.
• Feb. 2024 - Oct. 2024
  Google, Tokyo, Research Scientist
  Researching on speech processing.
• May 2023 - Aug. 2023
  Google, New York, Research Intern
  Researched on massive multilingual speech synthesis.
• Oct. 2022 - Jan. 2023
  Carnegie Mellon University, Visiting Scholar
  Researched on low-resource multilingual speech synthesis.
• Apr. 2022 - Sep. 2022
  Google, Tokyo, Student Researcher
  Researched on massive multilingual semi-supervised learning for speech synthesis.
• Mar. 2021 - Mar. 2022
  LINE Corporation, Part-Time Researcher
  Researched on noise-robust text-to-speech synthesis.
• Aug. 2021 - Sep. 2021
  Preferred Networks, Research Intern
  Worked on singing voice conversion.
• Aug. 2019 - Sep. 2019
  NEC Datascience Research Laboratories, Research Intern
  Worked on speech enhancement.
• Feb. 2019 - June 2019
  Recruit Holdings Co., Ltd., Engineering Intern & Part-Time Engineer
  Worked on data analysis and developed recommendation engine.

Reviewing

• IEEE ICASSP: 2023, 2024, 2025.
• INTERSPEECH: 2023, 2024, 2025.
• ACL, 2025.
• COLING, 2025.
• IEEE Signal Processing Letters: 2023, 2024, 2025.
• IEEE/ACM Transactions on Audio, Speech and Language Processing: 2023, 2024, 2025.

Misc.

• Aug. 2024
  Featured as Recommended Ph.D. Thesis in 2023 [link]
  Information Processing Society of Japan (IPSJ)
• July 2024
  Published review article on UTMOS (automatic speech quality assessment method) [link]
  The Journal of the Acoustical Society of Japan
• April 2024
  Research Talk at NII Yamagishi Lab.
  Multilingual Low-Resource Speech Synthesis Leveraging Self-Supervised/Unsupervised Learning
• June 2022
  Research Talk at Google Tokyo
  Self-Supervised Speech Resotoration for Historical Audio
• Jan. 2021
  Exhibition at Sainokuni Buisiness Arena 2021
  Research on Stress-Free, Real-Time, and Full-Band Voice Conversion Based on Perceptual Model
• Oct. 2020
  Exhibition at CEATEC 2020
  Research on Stress-Free, Real-Time, and Full-Band Voice Conversion Based on Perceptual Model