On the use of MFCC and SWT-based features for offensive speech detection in social media.
Safa ChebbiSara SekkateSofia Ben JebaraAbdellah AdibPublished in: INISTA (2021)
Keyphrases
- social media
- speech recognition
- feature extraction
- false positives
- feature set
- feature vectors
- speech signal
- mel frequency cepstral coefficients
- speaker recognition
- speaker identification
- cepstral coefficients
- audio features
- object detection
- low level
- detection method
- noisy environments
- image features
- multiscale
- feature selection
- gaussian mixture model
- extracted features
- wavelet transform
- social media data
- object recognition
- speech recognition systems
- feature space