Annotation Confidence vs. Training Sample Size: Trade-Off Solution for Partially-Continuous Categorical Emotion Recognition.
Elena RyuminaOxana VerkholyakAlexey KarpovPublished in: Interspeech (2021)
Keyphrases
- sample size
- emotion recognition
- number of training samples
- model selection
- upper bound
- worst case
- audio visual
- human computer interaction
- optimal solution
- progressive sampling
- emotional speech
- supervised learning
- facial expressions
- sentiment analysis
- information fusion
- user interface
- image retrieval
- training set
- metadata
- data sets