Macro-block dropout for improved regularization in training end-to-end speech recognition models.
Chanwoo KimSathish IndurtiJinhwan ParkWonyong SungPublished in: CoRR (2022)
Keyphrases
- video data
- end to end
- speech recognition
- motion vectors
- acoustic models
- macroblock
- hidden markov models
- automatic speech recognition
- speech recognizer
- language model
- congestion control
- pattern recognition
- speech recognition systems
- speech synthesis
- error concealment
- speech signal
- selection algorithm
- video coding
- speaker independent
- isolated word
- speaker identification
- rate distortion
- bit rate
- neural network