Transductive Parameter Transfer, Bags of Dense Trajectories and MILES for No-Audio Multimodal Speech Detection.
Laura Cabrera QuirosEkin GedikHayley HungPublished in: MediaEval (2018)
Keyphrases
- audio visual
- multi modal
- multimedia
- multi stream
- audio stream
- broadcast news
- emotion recognition
- multimodal fusion
- text to speech
- audio signals
- detection method
- cross modal
- visual speech
- story segmentation
- detection algorithm
- speech processing
- visual information
- digital audio
- noisy environments
- voice activity detection
- multimodal interfaces
- prosodic features
- cepstral features
- moving objects
- signal processing
- speaker identification
- multiple instance learning
- semi supervised
- acoustic features
- object detection
- speech recognition
- audio features
- audio recordings
- text classification
- human computer interaction
- training data
- hidden markov models
- semi supervised learning
- audio video
- transfer learning
- visual data
- multimodal interaction
- transductive learning
- trajectory data