Login / Signup

Weakly-supervised forced alignment of disfluent speech using phoneme-level modeling.

Theodoros KouzelisGeorgios ParaskevopoulosAthanasios KatsamanisVassilis Katsouros
Published in: CoRR (2023)
Keyphrases
  • weakly supervised
  • speech recognition
  • topic models
  • object class
  • semi supervised
  • superpixels
  • relation extraction
  • viewpoint
  • hidden markov models
  • d objects
  • object detection
  • co occurrence
  • higher order
  • visual features