Learning When to Speak: Latency and Quality Trade-offs for Simultaneous Speech-to-Speech Translation with Offline Models.
Liam DuganAnshul WadhawanKyle SpenceChris Callison-BurchMorgan McGuireVictor B. ZordanPublished in: CoRR (2023)
Keyphrases
- trade off
- language acquisition
- speech recognition
- learning process
- prior knowledge
- learning algorithm
- speech signal
- learned models
- high quality
- online learning
- real time
- learning systems
- learning problems
- generative model
- learning models
- language learning
- reinforcement learning
- information retrieval
- neural network
- computational model
- computational models
- knowledge acquisition
- dialogue system