Toranosuke Manabe

Toranosuke Manabe  真鍋 虎之祐

Ph.D. Student @ Keio University

I am a first-year Ph.D. student at Aoki Lab, Keio University, advised by Prof. Yoshimitsu Aoki.

I am working on multimodal learning, especially with human motion and speech. Currently, I am focusing on co-speech gesture generation.

More broadly, I am interested in representation learning, reinforcement learning, and human motion generation.

Email Scholar GitHub LinkedIn

Publications

Bold = me. * = equal contribution.

ProLAP figure
Interspeech 2026 🇦🇺
ProLAP: Probabilistic Language-Audio Pre-Training
Toranosuke Manabe, Yuchi Ishikawa, Hokuto Munakata, Yoshimitsu Aoki, Tatsuya Komatsu
arXiv
Sign-to-Speech Prosody Transfer figure
ICPR 2026 (Oral) 🇫🇷🏆 Best Scientific Paper Award
Sign-to-Speech Prosody Transfer via Sign Reconstruction-based GAN
Toranosuke Manabe, Yuto Shibata, Shinnosuke Takamichi, Yoshimitsu Aoki
Project arXiv Code
Listening without Looking figure
ICIP 2026 (Oral) 🇫🇮
Listening without Looking: Modality Bias in Audio-Visual Captioning
Yuchi Ishikawa, Toranosuke Manabe, Tatsuya Komatsu, Yoshimitsu Aoki
arXiv

Awards

Aug 2026
IMDS Best Scientific Paper Award (Track 3: Image, Speech, Signal and Video Processing) at ICPR 2026 (The 28th International Conference on Pattern Recognition).
"Sign-to-Speech Prosody Transfer via Sign Reconstruction-based GAN"
Aug 2025
MIRU Student Encouragement Award at MIRU 2025 (The 28th Meeting on Image Recognition and Understanding).
Mar 2025
Outstanding Bachelor's Thesis Presentation Award, Department of Electronics and Electrical Engineering, Keio University.