Speech Emotion Recognition Based on Self-Attention Weight Correction for Acoustic and Text Features.
Jennifer SantosoTakeshi YamadaKenkichi IshizukaTaiichi HashimotoShoji MakinoPublished in: IEEE Access (2022)
Keyphrases
- speech emotion recognition
- additional features
- feature extraction
- information retrieval
- feature vectors
- textual features
- structural features
- automatically extracted
- co occurrence
- text mining
- structural information
- classification accuracy
- speech recognition systems
- database
- semantic information
- low level
- object recognition
- feature weights
- feature space
- lexical features
- emotional speech