Leveraging Self-Supervised Speech Representations for Domain Adaptation in Speech Enhancement.
Ching Hua LeeChouchang YangRakshith Sharma SrinivasaYashas Malur SaiduttaJaejin ChoYilin ShenHongxia JinPublished in: ICASSP (2024)
Keyphrases
- speech enhancement
- domain adaptation
- noisy environments
- noise reduction
- speech signal
- signal to noise ratio
- single channel
- vocal tract
- linear prediction
- labeled data
- noisy speech
- multiple sources
- cross domain
- semi supervised
- wiener filter
- semi supervised learning
- target domain
- multi channel
- background noise
- speech recognition
- smoothing algorithm
- test data
- transfer learning
- image coding
- unlabeled data
- sentiment classification
- data sets
- co training
- document classification
- hidden markov models
- pairwise