SGEM: Test-Time Adaptation for Automatic Speech Recognition via Sequential-Level Generalized Entropy Minimization.
Changhun KimJoonhyung ParkHajin ShimEunho YangPublished in: CoRR (2023)
Keyphrases
- automatic speech recognition
- entropy minimization
- speech recognition
- speaker adaptation
- speech signal
- hidden markov models
- conversational speech
- noisy environments
- word recognition
- broadcast news
- word error rate
- speech retrieval
- multiscale
- spoken words
- haplotype inference
- recognition errors
- pattern recognition
- image processing
- acoustic features
- handwriting recognition
- wavelet transform