The training of the tone of Mandarin two-syllable words based on pitch projection synthesis speech.
Yanlu XieBei ZhangJinsong ZhangPublished in: ISCSLP (2014)
Keyphrases
- prosodic features
- spoken document retrieval
- speech recognition
- spontaneous speech
- n gram
- speech synthesis
- broadcast news
- text to speech
- text recognition
- hearing impaired
- speaker verification
- speech recognition systems
- continuous speech recognition
- speech recognition errors
- grapheme to phoneme conversion
- emotion recognition
- synthesized speech
- speech corpus
- automatic transcription
- automatic speech recognition
- training examples
- speech signal
- training process
- supervised learning
- fundamental frequency
- training corpus
- acoustic features
- audio visual
- recognition errors
- training set
- acoustic models
- speech sounds
- out of vocabulary
- spoken words
- formant frequencies
- conversational speech
- speaker independent
- vocal tract
- neural network
- human machine interaction
- spoken language
- noisy environments
- word sense disambiguation
- probabilistic model
- hidden markov models
- natural language