Global Syllable Vectors for Building TTS Front-End with Deep Learning.

Jinfu Ni Yoshinori Shiga Hisashi Kawai

Published in: INTERSPEECH (2017)

Keyphrases

deep learning
machine learning
unsupervised learning
unsupervised feature learning
text to speech
prosodic features
mental models
restricted boltzmann machine
feature vectors
weakly supervised
training data
deep architectures
information retrieval
similarity measure