DIRAC
Official name: Detection and Identification of Rare Audio-visual Cues
Official website: www.diracproject.org
Abstract:
The project aims at the design and development of an environment-adaptive autonomous active cognitive system that will detect and identify low prior probability unexpected events from the information derived by multiple, active information-seeking sensors. The system will probe for relevant cues, will autonomously adapt to new and changing environments and will reliably discard non-informative data.
BUT research task:
BUT’s task in DIRAC is to identify and describe unexpected information sources in human speech communication, develop basic approaches to identifying disagreements between data and prior expectations in audio. The group will focus on unexpected words, out-of-vocabulary (OOV) words: BUT researchers develop and test techniques for their identification, description (with subsequent ‘re-injection’ in the recognition system) and OOV clustering. Have a look at our DIRAC demo: www.prednasky.com/oov-fisher/
| Principal Investigator: | Hynek Heřmanský |
|---|---|
| Co-investigators: | Lukáš Burget, Mirko Hannemann, Stefan Kombrink, Tomáš Mikolov |
Grant agency:
DIRAC is a European project co-funded by the European Commission within the 6th Framework Programme under project number: EU FP6 IST – 027787 and runs from 01/2006 to 12/2010.
Logo:
