Table 7 MOS scores for all four voices for different amounts of data pruning
From: Developing a unit selection voice given audio without corresponding text
Percentage units used | Mean opinion score | |||
|---|---|---|---|---|
| Â | Test data = Olive | Test data = lecture | ||
| Â | ASR trained on Olive | ASR trained on LibriSpeech | ASR trained on lecture | ASR trained on LibriSpeech |
100 | 3.49 | 3.52 | 3.18 | 2.91 |
| Â | 3.51 | 3.47 | 3.21 | 3.22 |
50 | 3.28 | 3.22 | 2.99 | 3.05 |
30 | 3.08 | 3.00 | 2.90 | 2.93 |