Using the Bag-of-Audio-Word Feature Representation of ASR DNN Posteriors for Paralinguistic Classification.
Gábor GosztolyaPublished in: INTERSPEECH (2019)
Keyphrases
- feature representation
- feature extraction
- feature set
- feature representations
- face recognition
- classification accuracy
- low dimensional
- feature space
- machine learning
- feature vectors
- feature descriptors
- gaussian mixture model
- speech recognition
- co occurrence
- nearest neighbor
- training set
- data analysis
- object recognition
- image sequences
- decision trees
- image processing