Polyphone disambiguation and accent prediction using pre-trained language models in Japanese TTS front-end.
Rem HidaMasaki HamadaChie KamadaEmiru TsunooToshiyuki SekiyaToshiyuki KumakuraPublished in: CoRR (2022)
Keyphrases
- language model
- pre trained
- speech recognition
- language modeling
- document retrieval
- n gram
- query expansion
- language modelling
- probabilistic model
- context sensitive
- information retrieval
- prediction accuracy
- statistical language models
- retrieval model
- training data
- test collection
- document ranking
- co occurrence
- natural language
- smoothing methods
- control signals
- language models for information retrieval
- speech signal
- natural language processing
- wordnet
- generative model
- hidden markov models
- decision trees
- neural network
- data sets