Speech Enhancement with Weakly Labelled Data from AudioSet.
Qiuqiang KongHaohe LiuXingjian DuLi ChenRui XiaYuxuan WangPublished in: Interspeech (2021)
Keyphrases
- labelled data
- speech enhancement
- unlabelled data
- noisy environments
- noise reduction
- signal to noise ratio
- single channel
- speech signal
- linear prediction
- training data
- smoothing algorithm
- background noise
- wiener filter
- vocal tract
- multi channel
- additive noise
- concept drift
- speech recognition
- automatic speech recognition
- non stationary
- signal processing
- principal component analysis
- multiscale
- data sets