An application of minimum classification error to feature space transformations for speech recognition.
Ángel de la TorreAntonio M. PeinadoAntonio J. RubioVictoria E. SánchezJesús Esteban Díaz VerdejoPublished in: Speech Commun. (1996)
Keyphrases
- speech recognition
- feature space
- minimum classification error
- hidden markov models
- discriminative training
- automatic speech recognition
- statistical pattern recognition
- pattern recognition
- speech synthesis
- language model
- classification error
- training set
- speech recognizer
- feature vectors
- classification accuracy
- speaker identification
- speech signal
- speech recognition technology
- high dimensional
- data points
- principal component analysis
- kernel function
- feature selection
- maximum likelihood estimation
- dimensionality reduction
- gaussian mixture
- training samples
- speech recognition systems
- gaussian mixture model
- low dimensional
- feature set
- noisy environments
- feature extraction
- speaker independent
- kernel pca
- computer vision
- mean shift