On the Importance of Super-Gaussian Speech Priors for Machine-Learning Based Speech Enhancement.
Robert RehrTimo GerkmannPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2018)
Keyphrases
- speech enhancement
- machine learning
- noisy environments
- noise reduction
- speech signal
- signal to noise ratio
- single channel
- vocal tract
- linear prediction
- noisy speech
- speech recognition
- prior information
- pattern recognition
- multi channel
- smoothing algorithm
- edge preserving
- computer vision
- background noise
- wiener filter
- image coding
- additive noise
- feature selection
- image acquisition
- speaker identification
- gaussian mixture model
- probabilistic model
- bayesian networks
- feature extraction
- information retrieval