ALBAYZÍN evaluation challenges focus on evaluating different audio-visual technologies over TV and radio broadcast, and Basque Parliament content. Five evaluations are proposed:
- Speech to Text Challenge (S2TC), organized by RTVE and Universidad de Zaragoza, consists of automatically transcribe different types of TV shows.
- Speaker Diarization and Identity Asignement (SDIAC), organized by RTVE and Universidad de Zaragoza, consists of segmenting broadcast audio documents according to different speakers, linking those segments which originate from the same speaker and identify a closed set of speakers.
- Multimodal Diarization and Scene Description Challenge (MDSDC), organized by RTVE and Universidad de Zaragoza, consists of segmenting broadcast audio-visual documents according to a closed set of speakers, faces and scene descriptors and linking those segments which originate from the same speaker, face and scene descriptor.
- Search on Speech Challenge (SoSC), organized by Universidad San Pablo-CEU and AuDIaS from Universidad Autónoma de Madrid, consists of searching in audio content a list of terms/queries.
Text-to-Speech Alignment Evaluation (S2TAC), organized by University of the Basque Country (UPV/EHU), consists of aligning text and audio extracted from a plenary session of the Basque Parliament (bilingual audio).(THIS EVALUATION HAS BEEN CANCELLED)
More information about the different challenges, databases and online registration can be found in http://catedrartve.unizar.es/albayzin2020.html