Retrieving Speech Samples with Similar Emotional Content Using a Triplet Loss Function.
John B. HarvillMohammed Abdel-WahabReza LotfianCarlos BussoPublished in: ICASSP (2019)
Keyphrases
- loss function
- pairwise
- support vector
- learning to rank
- logistic regression
- speech recognition
- emotion recognition
- hinge loss
- regularization term
- risk minimization
- solution path
- emotional state
- training set
- reproducing kernel hilbert space
- stochastic gradient descent
- information retrieval
- convex loss functions
- boosting algorithms
- matrix factorization
- search engine