DDS: A new device-degraded speech dataset for speech enhancement.
Haoyu LiJunichi YamagishiPublished in: CoRR (2021)
Keyphrases
- speech enhancement
- noisy environments
- speech signal
- noise reduction
- signal to noise ratio
- vocal tract
- single channel
- noisy speech
- speech recognition
- linear prediction
- background noise
- smoothing algorithm
- wiener filter
- additive noise
- speech synthesis
- edge preserving
- multi channel
- non stationary
- feature set
- feature extraction
- speaker identification
- motion blur
- image restoration