Statistical modeling of speech feature vector trajectories based on a piecewise continuous mean path.
Mark M. ThomsonPublished in: ICASSP (1995)
Keyphrases
- statistical modeling
- piecewise continuous
- feature vectors
- statistical models
- speech recognition
- feature space
- euclidean distance
- predictive modeling
- speech segments
- feature set
- feature extraction
- collision free
- support vector machine
- shortest path
- configuration space
- speech synthesis
- image features
- nonparametric bayesian
- audio visual
- similarity measure
- moving objects
- spoken language
- speech signal
- trajectory data
- speaker identification
- text to speech
- gaussian mixture model
- data mining
- higher order
- least squares
- classification accuracy
- emotional state
- motion patterns
- statistical model
- broadcast news
- svm classifier
- dimensionality reduction
- mel frequency cepstral coefficients
- recognition engine
- pairwise
- feature selection