Skip to main content
Figure 10 | SpringerPlus

Figure 10

From: On pre-image iterations for speech enhancement

Figure 10

Results of PI with frequency-dependent determination of the kernel variance based on the SNR (PIDFSNR), with additional variation of the feature extraction parameters (PIDFSNR-Var), of the generalized subspace method (Subspace), spectral subtraction (SpecSub), and the MMSE log-spectral amplitude estimator (logMMSE) in terms of PEASS scores on the test set of the Noizeus database corrupted by car noise.

Back to article page