CSR: Speaker Recognition from Compressed VoIP Packet Stream.
Charu C. AggarwalDavid P. OlshefskiDebanjan SahaZon-Yin ShaePhilip S. YuPublished in: ICME (2005)
Keyphrases
- speaker recognition
- packet loss
- gaussian mixture model
- vector quantization
- speaker verification
- speaker identification
- probabilistic neural network
- speech recognition
- quality of service
- video quality
- bit rate
- hidden markov models
- machine learning
- neural network
- pattern classification
- mixture model
- ip networks
- similarity measure
- image processing