Parameter-Efficient Conformers via Sharing Sparsely-Gated Experts for End-to-End Speech Recognition.
Ye BaiJie LiWenjing HanHao NiKaituo XuZhuo ZhangCheng YiXiaorui WangPublished in: INTERSPEECH (2022)
Keyphrases
- end to end
- speech recognition
- hidden markov models
- language model
- speech synthesis
- speech processing
- automatic speech recognition
- speech recognition technology
- speech recognizer
- congestion control
- pattern recognition
- speaker identification
- speech signal
- speech recognition systems
- noisy environments
- speaker independent
- speech recognizers
- feature vectors
- speech retrieval
- neural network
- isolated word