J-ToneNet: A Transformer-based Encoding Network for Improving Tone Classification in Continuous Speech via F0 Sequences.
Yi-Fen LiuXiang-Li LuPublished in: INTERSPEECH (2023)
Keyphrases
- classification accuracy
- machine learning
- automatic classification
- pattern recognition
- classification models
- decision rules
- support vector machine svm
- classification algorithm
- support vector machine
- training set
- classification systems
- feature extraction
- feature selection
- pattern classification
- complex networks
- network structure
- classification method
- image classification
- power system
- classification process
- hidden markov models
- feature vectors
- feature space
- distribution network
- radial basis function network
- fractal image encoding
- data sets
- variable length
- network model
- unsupervised learning
- text classification
- preprocessing
- neural network