Skip to main content

Table 2 Development and test term list characteristics for RTVE22 database

From: Whisper-based spoken term detection systems for search on speech ALBAYZIN evaluation challenge

Term list

Development

Test

#INL (occ.)

307 (1151)

188 (930)

#OOL (occ.)

91 (351)

72 (109)

#SINGLE (occ.)

380 (1280)

217 (985)

#MULTI (occ.)

18 (222)

43 (54)

  1. occ. number of occurrences (in brackets), INL in-language, OOL out-of-language, SINGLE single-word terms, MULTI multi-word terms. The term length in the development term list varies between 4 and 25 graphemes (single-word term length varies between 4 and 20 graphemes, and multi-word term length varies between 7 and 25 graphemes). The term length in the test term list varies between 3 and 28 graphemes (single-word term length varies between 3 and 12 graphemes, and multi-word term length varies between 8 and 28 graphemes)