PauseSpeech: Natural Speech Synthesis via Pre-trained Language Model and Pause-Based Prosody Modeling.
Ji-Sang HwangSang-Hoon LeeSeong-Whan LeePublished in: ACPR (1) (2023)
Keyphrases
- speech synthesis
- speech recognition
- language model
- pre trained
- language modeling
- document retrieval
- prosodic features
- probabilistic model
- n gram
- information retrieval
- query expansion
- test collection
- automatic speech recognition
- retrieval model
- ad hoc information retrieval
- text to speech
- mixture model
- smoothing methods
- relevance model
- speech signal
- hidden markov models
- context sensitive
- training examples
- statistical language modeling
- training data
- face recognition
- machine learning