From: A novel voice conversion approach using admissible wavelet packet decomposition
Score | MOS (speech quality) | ABX (speaker identity) |
---|---|---|
1 | Bad (imperfect to perceive) | Totally different |
2 | Poor (almost impossible to perceive) | Certainly not |
3 | Fair (sound perception is not perfect) | Possibly different |
4 | Very good (cell phone quality) | More or less the same |
5 | Outstanding (perfect to perceive) | Totally same |