A novel Bayesian framework for speech enhancement using speech presence uncertainty.
Qing WangChangchun BaoFeng BaoPublished in: ChinaSIP (2015)
Keyphrases
- bayesian framework
- speech enhancement
- noisy environments
- noise reduction
- speech signal
- signal to noise ratio
- single channel
- vocal tract
- generative model
- linear prediction
- prior knowledge
- noisy speech
- hyperparameters
- speech recognition
- smoothing algorithm
- maximum a posteriori
- wiener filter
- multi channel
- probability density function
- additive noise
- expectation maximization
- image acquisition
- image coding
- background noise
- speech synthesis
- image compression
- edge detection
- machine learning