NAIST Simultaneous Speech-to-speech Translation System for IWSLT 2023

20572

Knowledge Triples

Source Entity	Relationship	Target Entity
Incremental Text to Speech Module Method	consistsOf	Acoustic Model Material
Incremental Text to Speech Module Method	consistsOf	Japanese Pronunciation Estimation Model OtherEntity
Incremental Text to Speech Module Method	consistsOf	Neural Vocoder OtherEntity
Multilingual Speech Translation Model Method	enhancedBy	Bilingual Prefix Alignment Method