Table 2 The distribution of the dataset used in our experiment
From: Exploiting contextual information for prosodic event detection using auto-context
| Â | F1 | F2 | M1 | M2 |
|---|---|---|---|---|
Number of utterances | 74 | 166 | 72 | 51 |
Number of sentences | 279 | 1,176 | 391 | 209 |
Number of words | 3,993 | 12,060 | 5,059 | 3,608 |
Number of syllables | 6,580 | 20,836 | 8,168 | 5,915 |
Number of accents | 2,253 | 7,063 | 2,564 | 1,933 |
Number of boundaries | 977 | 3,702 | 1,092 | 882 |
Accent occupancy (%) | 34.3 | 33.9 | 31.4 | 32.7 |
Boundary occupancy (%) | 14.9 | 17.8 | 13.4 | 14.9 |