Prosody Aware Word-Level Encoder Based on BLSTM-RNNs for DNN-Based Speech Synthesis.
Yusuke IjimaNobukatsu HojoRyo MasumuraTaichi AsamiPublished in: INTERSPEECH (2017)
Keyphrases
- speech synthesis
- word level
- speech recognition
- recurrent neural networks
- text to speech
- language independent
- document images
- machine translation
- document analysis
- prosodic features
- n gram
- viterbi algorithm
- word recognition
- character recognition
- hidden markov models
- word segmentation
- neural network
- semantic roles
- sentence level
- text mining
- cross lingual
- pattern recognition
- computer vision
- data mining