Research Promotion
- Construction of a scalable speech recognition system that automatically adapts according to recognition task
- Investment: Hoso Bunka Foundation
- Category: Technological Development Grant in 2016 (delegate: Nankaku)
- Term: Apr.2017 - Mar.2018
- Research on content generation by users using singing voice synthesis system based on statistical model
- Investment: Hoso Bunka Foundation
- Category: Technological Development Grant in 2015 (delegate: Tokuda)
- Term: Apr.2016 - Mar.2017
- A singing voice synthesis system based on a statistical model that learns and reproduces the voice quality and singing voice of a singer automatically
- A robust image recognition model to geometric variation based on Bayesian criterion
- Research on Generation Method of Personal Protection Sound Enabling Privacy Protection by Anonymizing Voice
- Construction of generic image model based on statistical approach
- Establishment of technical and social infrastructure for constructing various speech synthesis systems based on huge data
- Construction of technical and social infrastructure for constructing various speech synthesis systems based on huge data
- Reserch about the improvement of intelligibility in fast-talking Speech Synthesis based on Hidden Markov Model
- Voice system that automatically learn and reproduce how to sing and voice quality of the singer
- Bayesian factor analysis representing various voice characteristics for Speech Synthesis
- Research on next generation speech technology base for realizing speech ubiquitous information environment
- A statistical approach to construct generic image models
- Flexible speech synthesis based on the formulation of cross-language prosody
- A statistical approach to construct generic image models
- Japan-Finland International Bilateral Seminar
- Expressive multilingual Speech Synthesis
- Various technology of Speech Synthesis to produce digital media contents
- The realization of high efficient parallel Large Vocabulary Continuous Speech Recognition based on Dynamic Constrained Search Strategy
- Trajectory models for Speech Processing
- Research for improvement of Speech Synthesis based on HMM
- Speech Synthesis using Hidden Markov Model --The aim at developing emotional computer--
- Research for Speech Recognition using mechanism based on Kernel
- Research for Speech Recognition using mechanism based on Kernel
- Pioneering research on Voice Personal Authentication based on Minimum Classification Error Criterion and Mixtures of Factor Analyzers
- Speech Synthesis based on Hidden Markov Model --The realization variety of voice quality and speaking style--
- Speech Synthesis based on Hidden Markov Model --The aim at developing computer talk like human--
- Speech Synthesis based on Hidden Markov Model --The aim at developing computer talk like human--
- Development of Speech Synthesis System based on Hidden Markov Model.Development of interface of Multi-Modal Speech Recognition and Synthesis
- Speech Synthesis using Hidden Markov Model --The aim at developing emotional computer--
- Speech Synthesis using Hidden Markov Model --The aim at developing emotional computer--
- Vector Quantization based on statistics of vectors --Application to low bit rate speech and image coding--
- Speech Synthesis using Hidden Markov Model --The aim at developing emotional computer--
- Speech Synthesis based on Hidden Markov Model --The aim at developing emotional computerて--
- Speech Synthesis using Hidden Markov Model --The aim at developing machine talk like human--
- Development of low bit rate speech coding system using Mel-Generalized Cepstrum
- Mel-Cepstral Analysis of Speech and its application to Adaptive Signal Processing
Attending international conferences Promotion and another Promotion
- Investment: Tokyo Institude of Technology
- Category: Tokyo Institute of Technology The Centennial Memorial International Academic exchange fund(first class) (delegate:Tokuda)
- Year: 1992