Publications

Journal Paper

• Takaaki Saeki, Soumi Maiti, Xinjian Li, Shinji Watanabe, Shinnosuke Takamichi, Hiroshi Saruwatari
  Text-Inductive Graphone-Based Language Adaptation for Low-Resource Speech Synthesis
  IEEE/ACM Transactions on Audio, Speech and Language Processing, 2024.
  [ieee xplore]
• Takaaki Saeki, Shinnosuke Takamichi, Tomohiko Nakamura, Naoko Tanji, and Hiroshi Saruwatari
  SelfRemaster: Self-Supervised Speech Restoration for Historical Audio Resources
  IEEE Access, 2023.
  [ieee xplore] [github]
• Takaaki Saeki, Yuki Saito, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Real-Time Full-Band Voice Conversion with Sub-Band Modeling and Data-Driven Phase Estimation of Spectral Differentials
  IEICE Transactions on Information and Systems, 2021.
  [paper] [jstage] [demo]
  (Best Paper Award)
• Takaaki Saeki, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Incremental Text-to-Speech Synthesis Using Pseudo Lookahead with Large Pretrained Language Model
  IEEE Signal Processing Letters, 2021.
  [ieee xplore] [arxiv] [demo] [poster]
  (The 37th Telecom System Technology Award for Students from TAF)
  

Peer-Reviewed Conference Paper

• Takaaki Saeki, Soumi Maiti, Shinnosuke Takamichi, Shinji Watanabe, Hiroshi Saruwatari
  SpeechBERTScore: Reference-Aware Automatic Evaluation of Speech Generation Leveraging NLP Evaluation Metrics
  Annual Conference of the International Speech Communication Association (Interspeech), 2024.
  [arxiv] [github]
• Takaaki Saeki, Gary Wang, Nobuyuki Morioka, Isaac Elias, Kyle Kastner, Andrew Rosenberg, Bhuvana Ramabhadran, Heiga Zen, Françoise Beaufays, Hadar Shemtov
  Extending Multilingual Speech Synthesis to 100+ Languages without Transcribed Data
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.
  [arxiv] [demo] [slide]
• Kentaro Seki, Shinnosuke Takamichi, Takaaki Saeki, and Hiroshi Saruwatari
  Diversity-based core-set selection for text-to-speech with linguistic and acoustic features
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.
  [arxiv]
• Xinjian Li, Shinnosuke Takamichi, Takaaki Saeki, William Chen, Sayaka Shiota, Shinji Watanabe
  YODAS: Youtube-Oriented Dataset for Audio and Speech
  IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2023.
  [ieee xplore]
• Yuta Matsunaga, Takaaki Saeki, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Improving Robustness of Spontaneous Speech Synthesis with Linguistic Speech Regularization and Pseudo-Filled-Pause Insertion
  ISCA Speech Synthesis Workshop (SSW), 2023.
  [openreview]
• Takaaki Saeki, Soumi Maiti, Xinjian Li, Shinji Watanabe, Shinnosuke Takamichi, Hiroshi Saruwatari
  Learning to Speak from Text: Zero-Shot Multilingual Text-to-Speech with Unsupervised Text Pretraining
  International Joint Conference on Artificial Intelligence (IJCAI), 2023.
  [ijcai_proceedings] [arxiv] [demo] [github] [slide]
• Soumi Maiti, Yifan Peng, Takaaki Saeki, Shinji Watanabe
  SpeechLMScore: Evaluating Speech Generation Using Speech Language Model
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.
  [arxiv] [github]
• Dong Yang, Tomoki Koriyama, Yuki Saito, Takaaki Saeki, Detai Xin, and Hiroshi Saruwatari
  Duration-Aware Pause Insertion Using Pre-trained Language Model for Multi-speaker Text-to-Speech
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.
  [arxiv] [demo]
• Kentaro Seki, Shinnosuke Takamichi, Takaaki Saeki, and Hiroshi Saruwatari
  Text-to-Speech Synthesis from Dark Data with Evaluation-in-the-Loop Data Selection
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023.
  [arxiv]
  (IEEE SPS Travel Grant for IEEE ICASSP 2023)
• Takaaki Saeki, Heiga Zen, Zhehuai Chen, Nobuyuki Morioka, Gary Wang, Yu Zhang, Ankur Bapna, Andrew Rosenberg, and Bhuvana Ramabhadran
  Virtuoso: Massive Multilingual Speech-Text Joint Semi-Supervised Learning for Text-to-Speech
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2023. (Oral)
  [arxiv] [demo] [slide]
• Yoshifumi Nakano, Takaaki Saeki, Shinnosuke Takamichi, Katsuhito Sudoh, and Hiroshi Saruwatari
  vTTS: Visual-Text to Speech
  IEEE Spoken Language Technology Workshop (SLT), 2023.
  [arxiv] [corpus] [github]
• Yuta Matsunaga, Takaaki Saeki, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Empirical Study Incorporating Linguistic Knowledge on Filled Pauses for Personalized Spontaneous Speech Synthesis
  Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), 2022.
  [corpus] [demo] [github]
• Takaaki Saeki, Shinnosuke Takamichi, Tomohiko Nakamura, Naoko Tanji, and Hiroshi Saruwatari
  SelfRemaster: Self-Supervised Speech Restoration with Analysis-by-Synthesis Approach Using Channel Modeling
  Annual Conference of the International Speech Communication Association (Interspeech), 2022. (Oral)
  [arxiv] [demo] [github] [slide]
  (Google East Asia Student Travel Grants)
• Takaaki Saeki, Kentaro Tachibana, and Ryuichi Yamamoto
  DRSpeech: Degradation-Robust Text-to-Speech Synthesis with Frame-Level and Utterance-Level Acoustic Representation Learning
  Annual Conference of the International Speech Communication Association (Interspeech), 2022.
  [arxiv] [demo]
• Takaaki Saeki*, Detai Xin*, Wataru Nakata*, Tomoki Koriyama, Shinnosuke Takamichi, and Hiroshi Saruwatari
  (*Equal contribution)
  UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022
  Annual Conference of the International Speech Communication Association (Interspeech), 2022. (Oral)
  [arxiv] [github] [demo]
  (Ranked 1st Place in 10/16 Metrics)
• Yuta Matsunaga, Takaaki Saeki, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Personalized Filled-Pause Generation with Group-Wise Prediction Models
  Language Resources and Evaluation Conference (LREC), 2022.
  [paper] [arxiv] [github]
• Naoki Kimura, Zixiong Su, Takaaki Saeki and Jun Rekimoto
  SSR7000: A Synchronized Corpus of Ultrasound Tongue Imaging for End-to-End Silent Speech Recognition
  Language Resources and Evaluation Conference (LREC), 2022.
  [paper] [github]
• Takaaki Saeki, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Low-Latency Incremental Text-to-Speech Synthesis with Distilled Context Prediction Network
  IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2021.
  [ieee xplore] [arxiv] [demo] [slide]
• Takaaki Saeki, Yuki Saito, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Real-Time, Full-Band, Online DNN-Based Voice Conversion System Using a Single CPU
  Annual Conference of the International Speech Communication Association (Interspeech), 2020. (Show & Tell)
  [paper] [video]
• Naoki Kimura, Zixiong Su, and Takaaki Saeki
  End-to-End Deep Learning Speech Recognition Model for Silent Speech Challenge
  Annual Conference of the International Speech Communication Association (Interspeech), 2020. (Show & Tell)
  [paper] [video]
• Takaaki Saeki, Yuki Saito, Shinnosuke Takamichi, and Hiroshi Saruwatari
  Lifter Training and Sub-Band Modeling for Computationally Efficient and High-Quality Voice Conversion Using Spectral Differentials
  IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2020.
  [ieee xplore] [arxiv] [slide] [video]

Preprint

• Shinnosuke Takamichi, Ludwig Kürzinger, Takaaki Saeki, Sayaka Shiota, and Shinji Watanabe
  JTubeSpeech: Corpus of Japanese Speech Collected from YouTube for Speech Recognition and Speaker Verification
  arXiv preprint, 2112.09323
  [arxiv] [github]
• Tomoki Hayashi, Ryuichi Yamamoto, Takenori Yoshimura, Peter Wu, Jiatong Shi, Takaaki Saeki, Yooncheol Ju, Yusuke Yasuda, Shinnosuke Takamichi, and Shinji Watanabe
  ESPnet2-TTS: Extending the Edge of TTS Research
  arXiv preprint, 2110.07840
  [arxiv] [demo] [github]