Extracting sub-glottal and Supra-glottal features from MFCC using convolutional neural networks for speaker identification in degraded audio signals.
Anurag ChowdhuryArun RossPublished in: IJCB (2017)
Keyphrases
- speaker identification
- speech signal
- audio signals
- mel frequency cepstral coefficients
- audio signal
- speech recognition
- speaker recognition
- automatic speech recognition
- convolutional neural networks
- noisy environments
- feature extraction
- non stationary
- hidden markov models
- feature vectors
- gaussian mixture model
- acoustic features
- audio features
- noisy images
- image features
- machine learning
- broadcast news
- feature space
- feature set
- sound source
- classification accuracy