Robust audio-visual speech synchrony detection by generalized bimodal linear prediction.
Kshitiz KumarJirí NavrátilEtienne MarcheretVit LibalGerasimos PotamianosPublished in: INTERSPEECH (2009)
Keyphrases
- linear prediction
- visual speech
- cepstral coefficients
- speech signal
- noisy environments
- image coding
- audio visual
- audio signal
- speech recognition
- lossless compression
- speaker identification
- hidden markov models
- prediction error
- audio signals
- information retrieval
- non stationary
- multimedia
- maximum likelihood
- language model
- broadcast news
- feature extraction
- feature selection