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: ICASSP (2023)
Keyphrases
- language understanding
- speech recognition
- contextual constraints
- automatic speech recognition
- spoken dialogue systems
- dialogue system
- natural language understanding
- speech synthesis
- information retrieval
- language processing
- spoken documents
- spontaneous speech
- dialogue management
- semantic interpretation
- speech sounds
- text to speech
- general knowledge
- spoken language
- free text
- computational models
- domain specific
- co occurrence
- text mining
- hidden markov models