RÚV-DI Speaker Diarization (2021-10-14)
Judy Y. Fong, Jón Guðnason · 2020
These are a set of speaker diarization recipes which depend on the speech toolkit Kaldi. There are two types of recipes here. First are recipes used for decoding unseen audio. The second type of recipes are for training diarization models on the Ruv-di data. This tool also lists the DER for the Ruv-di dataset on most of the recipes. All DERs within this tool have no unscored collars and include overlapping speech Þessi pakki inniheldur forskirftir fyrir samraeðugreind fyrir hugbunaðarumhverfið Kaldi. Pakkinn inniheldur tvaer tegundir af forskirftum. Annars vegar forskriftir sem greina samraeður i nýjum hljoðskram og hins vegar forskriftir til að þjalfa ný samraeðugreindarlikon með Ruv-di gagnasafninu. Hluti forskriftanna innihalda villutiðni (DER) fyrir Ruv-di gagnasettið.