Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022



Takaaki Saeki , Detai Xin , Wataru Nakata , Tomoki Koriyama , Shinnosuke Takamichi , Hiroshi Saruwatari

* Submitted to INTERSPEECH 2022 

   Access Paper or Ask Questions

DRSpeech: Degradation-Robust Text-to-Speech Synthesis with Frame-Level and Utterance-Level Acoustic Representation Learning



Takaaki Saeki , Kentaro Tachibana , Ryuichi Yamamoto

* Submitted to INTERSPEECH 2022 

   Access Paper or Ask Questions

SelfRemaster: Self-Supervised Speech Restoration with Analysis-by-Synthesis Approach Using Channel Modeling



Takaaki Saeki , Shinnosuke Takamichi , Tomohiko Nakamura , Naoko Tanji , Hiroshi Saruwatari

* Submitted to INTERSPEECH 2022 

   Access Paper or Ask Questions

Personalized filled-pause generation with group-wise prediction models



Yuta Matsunaga , Takaaki Saeki , Shinnosuke Takamichi , Hiroshi Saruwatari

* Submitted to LREC 2022 

   Access Paper or Ask Questions

JTubeSpeech: corpus of Japanese speech collected from YouTube for speech recognition and speaker verification



Shinnosuke Takamichi , Ludwig Kürzinger , Takaaki Saeki , Sayaka Shiota , Shinji Watanabe

* Submitted to ICASSP2022 

   Access Paper or Ask Questions

ESPnet2-TTS: Extending the Edge of TTS Research



Tomoki Hayashi , Ryuichi Yamamoto , Takenori Yoshimura , Peter Wu , Jiatong Shi , Takaaki Saeki , Yooncheol Ju , Yusuke Yasuda , Shinnosuke Takamichi , Shinji Watanabe

* Submitted to ICASSP2022. Demo HP: https://espnet.github.io/icassp2022-tts/ 

   Access Paper or Ask Questions

Low-Latency Incremental Text-to-Speech Synthesis with Distilled Context Prediction Network



Takaaki Saeki , Shinnosuke Takamichi , Hiroshi Saruwatari

* Accepted for ASRU2021 

   Access Paper or Ask Questions

Incremental Text-to-Speech Synthesis Using Pseudo Lookahead with Large Pretrained Language Model



Takaaki Saeki , Shinnosuke Takamichi , Hiroshi Saruwatari

* Submitted to IEEE Signal Processing Letters 

   Access Paper or Ask Questions