Table 5 Number of terms that are shared between databases and datasets
From: Whisper-based spoken term detection systems for search on speech ALBAYZIN evaluation challenge
Dataset | MAVIR dev. | MAVIR test | RTVE22 dev. | RTVE22 test | SPARL22 |
|---|---|---|---|---|---|
MAVIR dev. | – | 80 | 18 | 10 | 18 |
MAVIR test | 80 | – | 3 | 5 | 12 |
RTVE22 dev. | 18 | 3 | – | 25 | 15 |
RTVE22 test | 10 | 5 | 25 | – | 8 |
SPARL22 test | 18 | 12 | 15 | 8 | – |