Using representation learning and out-of-domain data for a paralinguistic speech task.
Benjamin MildeChris BiemannPublished in: INTERSPEECH (2015)
Keyphrases
- data sets
- database
- background knowledge
- image data
- supervised learning
- data analysis
- data distribution
- training data
- high quality
- data structure
- learning systems
- end users
- raw data
- original data
- human experts
- synthetic data
- input data
- small number
- data points
- data sources
- data model
- prior knowledge
- learning process
- reinforcement learning
- language acquisition
- knowledge discovery
- neural network
- data mining
- learning algorithm
- domain independent
- experimental data
- missing data
- high dimensional data
- image representation
- statistical analysis
- online learning