Deep Neural Network Based Discriminative Training for I-Vector/PLDA Speaker Verification.
Tieran ZhengJiqing HanGuibin ZhengPublished in: ICASSP (2018)
Keyphrases
- speaker verification
- discriminative training
- hidden markov models
- noisy environments
- probabilistic interpretation
- automatic speech recognition
- maximum likelihood
- neural network
- posterior probability
- multilayer perceptron
- audio visual
- support vector machine
- emotion recognition
- co training
- speech recognition
- language identification
- using artificial neural networks
- feature vectors
- data sets
- multi modal
- training data
- training set
- speech signal
- gaussian mixture model
- unsupervised learning