Comparison of I-vector and GMM-UBM approaches to speaker identification with TIMIT and NIST 2008 databases in challenging environments.
Musab T. S. Al-KaltakchiWai Lok WooSatnam Singh DlayJonathon A. ChambersPublished in: EUSIPCO (2017)
Keyphrases
- speaker identification
- gaussian mixture model
- speech recognition
- speaker recognition
- broadcast news
- speaker diarization
- noisy environments
- speech signal
- dynamic environments
- probabilistic neural network
- language identification
- hidden markov models
- feature extraction
- em algorithm
- neural network
- mel frequency cepstral coefficients
- speaker verification
- information retrieval systems
- probabilistic model
- multiscale
- image sequences