Employing both Gender and Emotion Cues to Enhance Speaker Identification Performance in Emotional Talking Environments.
Ismail ShahinPublished in: CoRR (2017)
Keyphrases
- speaker identification
- emotional state
- emotion recognition
- speech signal
- speech recognition
- gaussian mixture model
- noisy environments
- broadcast news
- feature extraction
- facial expressions
- affective states
- high level
- neural network
- mixture model
- automatic speech recognition
- visual features
- em algorithm
- classification accuracy
- probabilistic model
- feature selection
- machine learning