CTC Blank Triggered Dynamic Layer-Skipping for Efficient CTC-based Speech Recognition.
Junfeng HouPeiyao WangJincheng ZhangMeng YangMinwei FengJingcheng YinPublished in: CoRR (2024)
Keyphrases
- speech recognition
- hidden markov models
- language model
- speech recognizer
- speech synthesis
- speech processing
- speech understanding
- speech signal
- speech recognition technology
- speech recognition systems
- automatic speech recognition
- pattern recognition
- handwriting recognition
- noisy environments
- speech retrieval
- keyword spotting
- speaker identification
- neural network
- mixture model
- classification accuracy
- natural language
- speaker recognition
- text to speech
- em algorithm
- speaker dependent
- training data
- machine learning