Tackling Perception Bias in Unsupervised Phoneme Discovery Using DPGMM-RNN Hybrid Model and Functional Load.
Bin WuSakriani SaktiJinsong ZhangSatoshi NakamuraPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2021)
Keyphrases
- hybrid model
- recurrent neural networks
- artificial neural networks
- nearest neighbor
- support vector regression
- hybrid models
- back propagation neural network
- speech recognition
- load balancing
- support vector machine svm
- semi supervised
- supervised learning
- unsupervised learning
- visual perception
- context dependent
- discovery process
- machine learning
- load forecasting
- training data
- unsupervised manner
- hidden markov models
- pattern discovery
- data mining
- long term
- knowledge discovery
- feed forward
- fuzzy logic