Robust speaker recognition based on DNN/i-vectors and speech separation.
Jorge ChangDeLiang WangPublished in: ICASSP (2017)
Keyphrases
- speaker recognition
- gaussian mixture model
- speaker verification
- vector quantization
- speaker identification
- speech signal
- probabilistic neural network
- noisy environments
- speech recognition
- feature vectors
- mel frequency cepstral coefficients
- audio visual
- language identification
- image processing
- mixture model
- training data
- feature extraction