Soft-Target Training with Ambiguous Emotional Utterances for DNN-Based Speech Emotion Classification.
Atsushi AndoSatoshi KobashikawaHosana KamiyamaRyo MasumuraYusuke IjimaYushi AonoPublished in: ICASSP (2018)
Keyphrases
- emotion classification
- emotion recognition
- speech segments
- emotional speech
- training process
- emotional state
- audio visual
- lexical features
- syntactic analysis
- automatic speech recognition
- training set
- sentiment analysis
- information fusion
- human computer interaction
- speech recognition
- speaker verification
- facial expressions
- test set
- training data
- training examples
- training samples
- hidden markov models
- facial images
- sentence level
- multi modal
- supervised learning
- speaker independent
- feature space
- learning algorithm