Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 4 Worst-case scenario for a single CD codebook under traffic noise

From: Robust Bayesian estimation for context-based speech enhancement

 

ΔPESQ

ΔSSNR (in dB)

Input SSNR

0 dB

5 dB

10 dB

0 dB

5 dB

10 dB

CBSE with CD

−0.13±0.10

−0.02±0.11

−0.01±0.07

5.98±1.78

5.28±1.65

3.45±1.63

CBSE with CI

0.09±0.09

0.32±0.09

0.42±0.09

7.35±1.25

7.12±1.25

5.81±1.07

Proposed

0.08±0.09

0.29±0.07

0.39±0.08

7.34±1.24

7.05±1.22

5.75±1.07

Reference

0.21±0.08

0.28±0.10

0.34±0.09

3.21±0.65

3.65±0.65

3.35±0.58

  1. The CD codebook is modeling two aspects of context, speaker and acoustic environment. Both mean and standard deviation values are reported.
Back to article page