Table 6 Performance comparison of auto-context, acoustic context, and their combinations
From: Exploiting contextual information for prosodic event detection using auto-context
| Â | No | Acoustic | Auto-context | Combination |
|---|---|---|---|---|
| Â | context | context | Â | Â |
Two-way (%) | Â | Â | Â | Â |
Accuracy | 79.3 | 81.2 | 81.2 (M=3 itr = 3) | 82.0 (M=3, itr = 4) |
F-score | 66.7 | 70.8 | 71.1 (M=3 itr = 4) | 73.0 (M=3, itr = 4) |
Four-way (%) | Â | Â | Â | Â |
Accuracy | 74.2 | 76.3 | 75.7 (M=2 itr = 6) | 77.0 (M=2, itr = 1) |
F-score | 54.3 | 59.9 | 59.5 (M=2 itr = 6) | 63.0 (M=24, itr = 4) |