T5lephone: Bridging Speech and Text Self-supervised Models for Spoken Language Understanding via Phoneme level T5.
Chan-Jan HsuHo-Lam ChungHung-yi LeeYu TsaoPublished in: CoRR (2022)
Keyphrases
- language understanding
- speech recognition
- contextual constraints
- automatic speech recognition
- natural language understanding
- spoken dialogue systems
- speech synthesis
- dialogue system
- dialogue management
- general knowledge
- hidden markov models
- text to speech
- spoken language
- language processing
- computational models
- domain knowledge
- broadcast news
- speech signal
- free text
- computational model
- vocal tract
- spontaneous speech
- information retrieval
- speech sounds