A deep learning approach for robust speaker identification using chroma energy normalized statistics and mel frequency cepstral coefficients.
J. V. Thomas AbrahamA. Nayeemulla KhanA. ShahinaPublished in: Int. J. Speech Technol. (2023)
Keyphrases
- speaker identification
- deep learning
- mel frequency cepstral coefficients
- noisy environments
- speaker recognition
- speech recognition
- gaussian mixture model
- speech signal
- probabilistic neural network
- broadcast news
- feature extraction
- unsupervised learning
- noise reduction
- automatic speech recognition
- speaker verification
- machine learning
- mixture model
- spectral features
- neural network
- multiscale