Enhancement of Noisy Reverberant Speech Using Polynomial Matrix Eigenvalue Decomposition.
Vincent W. NeoChristine EversPatrick A. NaylorPublished in: IEEE ACM Trans. Audio Speech Lang. Process. (2021)
Keyphrases
- eigenvalue decomposition
- speech signal
- symmetric matrix
- blind source separation
- speech recognition
- least squares
- spectral clustering
- hidden markov models
- non stationary
- auto correlation
- low dimensional
- signal subspace
- feature space
- semidefinite programming
- pseudo inverse
- spectral analysis
- missing data
- image processing
- language model