OLKAVS: An Open Large-Scale Korean Audio-Visual Speech Dataset.
Jeongkyun ParkJung-Wook HwangKwanghee ChoiSeung-Hyun LeeJun Hwan AhnRae-Hong ParkHyung-Min ParkPublished in: CoRR (2023)
Keyphrases
- visual speech
- hidden markov models
- audio visual speech recognition
- visual speech recognition
- speaker identification
- audio signals
- multimedia
- audio signal
- speech signal
- noisy environments
- broadcast news
- lip reading
- acoustic features
- text to speech
- video signals
- pattern recognition
- automatic speech recognition
- multi stream
- audio visual
- feature set
- computational complexity