FeaRLESS: Feature Refinement Loss for Ensembling Self-Supervised Learning Features in Robust End-to-end Speech Recognition.
Szu-Jui ChenJiamin XieJohn H. L. HansenPublished in: CoRR (2022)
Keyphrases
- end to end
- speech recognition
- noisy environments
- pattern recognition
- feature vectors
- hidden markov models
- feature set
- language model
- cepstral coefficients
- speech signal
- speech recognition systems
- speech processing
- speech synthesis
- speech recognition technology
- speech retrieval
- speech recognizer
- word processing
- speaker recognition
- speaker identification
- automatic speech recognition