The use of discriminant neural networks in the integration of acoustic cues for voicing into a continuous-word recognition system.
Claude LefèbvreDariusz A. ZwierzynskiPublished in: ICSLP (1990)
Keyphrases
- neural network
- prosodic features
- feature extraction
- genetic algorithm
- pattern recognition
- co occurrence
- discriminant analysis
- n gram
- handwritten words
- high level
- fuzzy logic
- data integration
- word recognition
- recognition algorithm
- text to speech
- speaker verification
- recurrent neural networks
- multi layer
- training process
- activation function
- keywords
- self organizing maps
- fault diagnosis
- speech recognition systems
- artificial neural networks