PRODIS - a Speech Database and a Phoneme-based Language Model for the Study of Predictability Effects in Polish.
Zofia MaliszJan ForemskiMalgorzata KulPublished in: LREC/COLING (2024)
Keyphrases
- language model
- speech recognition
- probabilistic model
- word error rate
- n gram
- speech synthesis
- language modeling
- automatic speech recognition
- test collection
- language modelling
- speech signal
- hidden markov models
- document retrieval
- statistical language models
- search engine
- handwriting recognition
- context sensitive
- retrieval model
- error rate
- co occurrence
- clustering algorithm