Simultaneous-Speaker Voice Activity Detection and Localization Using Mid-Fusion of SVM and HMMs.
Vicente P. MinottoCláudio Rosito JungBowon LeePublished in: IEEE Trans. Multim. (2014)
Keyphrases
- voice activity detection
- speech recognition
- hidden markov models
- noisy environments
- support vector machine svm
- support vector
- automatic speech recognition
- support vector machine
- speaker verification
- speaker identification
- acoustic models
- knn
- speaker independent
- multi class
- speech recognizer
- pattern recognition
- speaker dependent
- speech signal
- svm classifier
- fusion method
- feature vectors
- information fusion
- localization algorithm
- multi sensor
- language model
- feature selection
- image fusion
- svm classification
- speaker recognition
- data fusion
- speaker diarization
- training data
- multi class classification
- discriminative training
- feature space
- multiscale
- machine learning
- audio visual
- generalization ability
- neural network
- visual information
- classification algorithm
- multi stream