HMM-based speech recognition using decision trees instead of GMMs.
Remco TeunenMasami AkaminePublished in: INTERSPEECH (2007)
Keyphrases
- decision trees
- gaussian mixture model
- hidden markov models
- speech recognition
- speaker recognition
- automatic speech recognition
- random forest
- predictive accuracy
- speaker identification
- decision tree induction
- audio visual
- speech signal
- speech synthesis
- machine learning algorithms
- rule induction
- decision rules
- naive bayes
- spoken dialogue systems
- emotion recognition
- dialogue system
- endpoint detection
- spoken language
- recognition engine
- rule sets
- multi modal
- machine learning
- emotional state
- ensemble methods
- em algorithm
- feature construction
- recognition rate
- decision tree algorithm
- pattern recognition
- training data
- text to speech
- decision tree learning
- feature selection
- neural network