Deep neural networks with auxiliary Gaussian mixture models for real-time speech recognition.
Xin LeiHui LinGeorg HeigoldPublished in: ICASSP (2013)
Keyphrases
- speech recognition
- gaussian mixture model
- speaker recognition
- speaker identification
- neural network
- pattern recognition
- mixture model
- language model
- mel frequency cepstral coefficients
- hidden markov models
- speech signal
- em algorithm
- automatic speech recognition
- speech synthesis
- speech recognition technology
- expectation maximization
- probability density function
- noisy environments
- maximum likelihood
- feature vectors
- speech recognizer
- probabilistic neural network
- speech recognition systems
- speaker independent
- bayesian information criterion
- speaker diarization
- machine learning
- gaussian distribution
- self organizing maps
- vector quantization
- feature extraction