Learning speaker, addressee and overlap detection models from multimodal streams.
Oriol VinyalsDan BohusRich CaruanaPublished in: ICMI (2012)
Keyphrases
- prior knowledge
- learning process
- learning tasks
- statistical models
- learning algorithm
- learning systems
- supervised learning
- learning models
- knowledge acquisition
- online learning
- structured prediction
- machine learning
- noisy environments
- discriminative learning
- accurate models
- audio visual
- multi modal
- learning problems
- real time
- video sequences
- multimedia
- data sets