Research on controllable and natural media synthesis
We aim to develop the technology to synthesize media (e.g., natural images and speech) in a controllable manner to make it possible to create or convert media as intended.
2012 The Hatakeyama Award, The Japan Society of Mechanical Engineers
2012 ICPR 2012 Best Student Paper Award, The 21st International Conference on Pattern Recognition (ICPR 2012)
2017 IEICE ISS Young Researcher's Award in Speech Field
2020 Dean's Award, Graduate School of Information Science and Technology, The University of Tokyo
Takuhiro Kaneko, Kaoru Hiramatsu, Kunio Kashino, “Generative Adversarial Image Synthesis with Decision Tree Latent Controller,” The 31st IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2018), June 2018.
Takuhiro Kaneko, Kaoru Hiramatsu, Kunio Kashino, “Generative Attribute Controller with Conditional Filtered Generative Adversarial Networks,” The 30th IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2017), July 2017.
Takuhiro Kaneko, Hirokazu Kameoka, Kou Tanaka, Nobukatsu Hojo, “MaskCycleGAN-VC: Learning Non-parallel Voice Conversion with Filling in Frames,” The 46th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2021), June 2021.
Takuhiro Kaneko, Hirokazu Kameoka, Kou Tanaka, Nobukatsu Hojo, “CycleGAN-VC3: Examining and Improving CycleGAN-VCs for Mel-spectrogram Conversion,” The 21st Annual Conference of the International Speech Communication Association (Interspeech 2020), Oct. 2020.
Image Synthesis, Speech Synthesis, Voice Conversion, Machine Learning, Deep Learning