OLKAVS: An Open Large-Scale Korean Audio-Visual Speech Dataset.
Jeongkyun ParkJung-Wook HwangKwanghee ChoiSeung-Hyeon LeeJun Hwan AhnRae-Hong ParkHyung-Min ParkPublished in: ICASSP (2024)
Keyphrases
- visual speech
- audio visual speech recognition
- hidden markov models
- visual speech recognition
- speaker identification
- audio signals
- lip reading
- noisy environments
- broadcast news
- audio signal
- multimedia
- multi stream
- video signals
- text to speech
- audio visual
- multi modal
- acoustic features
- speech signal
- image quality
- visual data
- visual information