Multi-speaker modeling and speaker adaptation for DNN-based TTS synthesis

• Published in: 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 4475 - 4479
• Main Authors: Yuchen Fan, Yao Qian, Soong, Frank K., Lei He
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2015
• Subjects: Acoustics; Adaptation models; deep neural networks; Hidden Markov models; multi-task learning; Pragmatics; Speech; statistical parametric speech synthesis; Training; Training data; transfer learning

In DNN-based TTS synthesis, DNNs hidden layers can be viewed as deep transformation for linguistic features and the output layers as representation of acoustic space to regress the transformed linguistic features to acoustic parameters. The deep-layered architectures of DNN can not only represent hi...