Speech enhancement based on hidden Markov model with discrete cosine transform coefficients using Laplace and Gaussian distributions.
Ali AroudiHadi VeisiHossein SametiPublished in: ISSPA (2012)
Keyphrases
- hidden markov models
- gaussian distribution
- transform domain
- transform coefficients
- speech enhancement
- speech signal
- noisy environments
- speech recognition
- spatial domain
- noise reduction
- signal to noise ratio
- single channel
- maximum likelihood
- frequency domain
- transform coding
- gaussian mixture model
- conditional random fields
- reconstructed image
- expectation maximization
- high resolution
- denoising
- image compression
- macroblock
- multi channel
- subband
- bit rate