Unified Speech-Text Pre-training for Speech Translation and Recognition.
Yun TangHongyu GongNing DongChanghan WangWei-Ning HsuJiatao GuAlexei BaevskiXian LiAbdelrahman MohamedMichael AuliJuan PinoPublished in: CoRR (2022)
Keyphrases
- recognition engine
- text recognition
- text to speech
- speech recognition
- text to speech synthesis
- hearing impaired
- speech corpus
- speech synthesis
- spoken words
- english text
- information retrieval
- automatic speech recognition systems
- speech signal
- speaker independent
- noisy environments
- hidden markov models
- speech sounds
- text input
- audio visual
- recognition accuracy
- recognition rate
- multi lingual
- character recognition
- pattern recognition
- object recognition
- speaker recognition
- feature extraction
- continuous speech recognition
- neural network
- conversational speech
- vocal tract
- speech recognizer
- broadcast news
- document analysis