| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 6951858 | Digital Signal Processing | 2018 | 9 Pages |
Abstract
- Mean square error does not directly align with the perceptual speech quality.
- The Mel-scale weight is proposed to emphasize critical frequency bands.
- The objective function incorporates temporal and spectral variation similarities.
Related Topics
Physical Sciences and Engineering
Computer Science
Signal Processing
Authors
Tae Gyoon Kang, Jong Won Shin, Nam Soo Kim,