Professor Akamine graduated from the Department of Electrical Engineering, Faculty of Science and Engineering, University of the Ryukyus in 1979, completed the Master's Program in Information Engineering, Graduate School of Engineering, Tohoku University in 1982, and completed the Doctoral Program at the Department of Electronic Engineering, Graduate School of Engineering, Tohoku University in 1985, and joined Tokyo Shibaura Electric Co., Ltd. (Present: TOSHIBA CORPORATION) in the same year. Since 2005, he has been the Senior Fellow of the Toshiba Corporate Research and Development Center, since 2016, he has also served as a project professor at the Graduate School of Science and Engineering, Keio University, and since 2019 he has been a research professor at the Graduate School of Engineering, Tohoku University. His achievements as a corporate researcher are remarkable, and he has been active in the field of electronics, information and communication, particularly in the research of practical applications of speech synthesis.He has led the research and development on speech synthesis for many years. He proposed a new paradigm of speech synthesis called closed loop training, and developed a speech synthesis method based on this for the first time in the world and made a great contribution to the dramatic improvement of performance and industrial development.
Until he proposed the new paradigm, in speech synthesis, parameters had been individually optimized through various intermediate signal processing from input to output. On the contrary, in the closed-loop training that he developed, by formulating the distortion between the final synthesized speech and the teaching signal as the evaluation quantity, it is possible to automatically learn the parameters so as to minimize the distortion. As a result, he succeeded not only in dramatically improving the quality of synthesized speech, but also in achieving a synthesis process that used to require several GB of memory with only a few hundred KB of memory.
This pioneering technology for speech synthesis has been widely adopted not only for Toshiba products but also for consumer equipment such as car navigation systems, electronic dictionaries, and video game machines, as well as social infrastructure equipment such as escalators and elevators, and has greatly contributed to the development of the industry. Even today, even if the speech synthesis method is replaced by a statistical model or machine learning methods, this closed-loop training paradigm has been used consistently and plays a central role in cloud services for speech synthesis (for example, Toshiba's RECAIUS(TM )) and smart speakers, which are growing rapidly.
At the IEICE, he has been a reviewer for many years and an associate editor of a special issue of the paper, and has been instrumental in managing and activating the Institute. He has also played the following important roles in various conferences related to technology promotion and academic promotion: Multi-media Committee Member of ITU-T’s Information and Communication Technology Subcommittee at Ministry of Internal Affairs and Communications from 1999 to 2011, Multi-media Application Special Committee Member of the Telecommunication Technology Committee from 2011 to 2018, Speech Resources Promotion Committee Member at National Institute of Informatics from 2015 to 2018, Steering Committee Member at the Research Institute of Electrical Communication, Tohoku University from 2016 to 2018, etc.
He received the IEICE Information and Systems Society Paper Award for his research on speech synthesis, and received the IEICE Achievement Award and Milestone for his contribution to "Pioneering research on high-quality speech synthesis methods suitable for memory saving.", and he made remarkable achievements in the field of speech synthesis and made a significant contribution to academic development. Also, due to these achievements, in addition to receiving a total of 11 awards, such as the Minister of Education, Culture, Sports, Science and Technology Award for Outstanding Research, the Ichimura Industry Award for outstanding achievement, and the Prime Minister's Invention Award for national invention, he has been awarded the Purple Ribbon as the leader in speech synthesis in Japan.
As mentioned above, in addition to his research achievements on speech synthesis technology, his contribution to the development of the IEICE and the electronic information communication field is extremely remarkable, and we are convinced that he is worthy of receiving the IEICE’s Distinguished Achievement and Contributions Award.
