Audio Imputation Using the Non-negative Hidden Markov Model.
Jinyu HanGautham J. MysoreBryan PardoPublished in: LVA/ICA (2012)
Keyphrases
- hidden markov models
- audio signals
- speech recognition
- multimedia
- multi stream
- missing values
- visual speech
- gesture recognition
- markov models
- missing data
- visual data
- speaker identification
- continuous hidden markov models
- visual information
- markov model
- sequential data
- audio visual
- conditional random fields
- automatic speech recognition
- viterbi algorithm
- speech signal
- hidden state
- hidden states
- phoneme recognition
- multi modal