Login / Signup

A dilemma of ground truth in noisy speech separation and an approach to lessen the impact of imperfect training data.

Matthew MaciejewskiJing ShiShinji WatanabeSanjeev Khudanpur
Published in: Comput. Speech Lang. (2023)
Keyphrases
  • ground truth
  • training data
  • high quality
  • training set
  • speech recognition
  • noisy speech
  • prior knowledge
  • neural network
  • hidden markov models
  • speech enhancement
  • multiresolution
  • automatic speech recognition systems