FAU IISAH Corpus – A German Speech Database Consisting of Human-Machine and Human-Human Interaction Acquired by Close-Talking and Far-Distance Microphones

8300

Knowledge Triples

Source Entity Relationship Target Entity
usedIn
partOf