RNN-based prosodic modeling for mandarin speech and its application to speech-to-text conversion.
Wern-Jun WangYuan-Fu LiaoSin-Horng ChenPublished in: Speech Commun. (2002)
Keyphrases
- speech recognition
- prosodic features
- speech synthesis
- text to speech
- speech signal
- nearest neighbor
- text to speech synthesis
- recurrent neural networks
- speaker verification
- hidden markov models
- database
- broadcast news
- emotion recognition
- pattern recognition
- speaker independent
- training data
- speech recognizer
- speech recognition systems
- case study
- feed forward
- feature vectors
- high dimensional
- neural network