Using genetic algorithms to weight acoustic features for speaker recognition.
Maider ZamalloaGermán BordelLuis Javier RodríguezMikel PeñagarikanoJuan Pedro UribePublished in: INTERSPEECH (2006)
Keyphrases
- speaker recognition
- acoustic features
- speaker verification
- mel frequency cepstral coefficients
- speech signal
- noisy environments
- speaker identification
- audio visual
- speech recognition
- genetic algorithm
- emotion recognition
- multilayer perceptron
- gaussian mixture model
- vector quantization
- automatic speech recognition
- face verification
- probabilistic neural network
- hidden markov models
- neural network
- using artificial neural networks
- multi modal
- image processing