Dynamic Encoder Size Based on Data-Driven Layer-wise Pruning for Speech Recognition.
Jingjing XuWei ZhouZijian YangEugen BeckRalf SchlüterPublished in: CoRR (2024)
Keyphrases
- speech recognition
- data driven
- hidden markov models
- language model
- speech signal
- speech synthesis
- automatic speech recognition
- speech processing
- pattern recognition
- noisy environments
- speech understanding
- speech recognition technology
- speech recognition systems
- speech recognizer
- speaker identification
- speech retrieval
- isolated word
- keyword spotting
- computer vision
- motion estimation
- image processing
- speech recognizers
- neural network