A Bayesian network approach combining pitch and spectral envelope features to reduce channel mismatch in speaker verification and forensic speaker recognition.
Mijail ArcienegaAnil AlexanderPhilipp ZimmermannAndrzej DrygajloPublished in: INTERSPEECH (2005)
Keyphrases
- speaker recognition
- speaker verification
- mel frequency cepstral coefficients
- gaussian mixture model
- acoustic features
- bayesian networks
- probabilistic neural network
- noisy environments
- vector quantization
- speaker identification
- feature vectors
- feature extraction
- low level
- multilayer perceptron
- audio visual
- extracting features
- spectral features
- multi modal
- language model
- neural network
- using artificial neural networks
- face verification
- image compression
- feature set
- probabilistic model
- feature space