2019/04/07 by Suwon Shon, Najim Dehak, Shon, Suwon +5 · 1 citation
Computer Science · #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Sound (cs.SD) #Speech Recognition and Synthesis #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1904.04240
openalex publication_date 2019/04/07 · openalex created_date 2022/07/24 · openalex updated_date 2026/07/28
The Multi-target Challenge aims to assess how well current speech technology\nis able to determine whether or not a recorded utterance was spoken by one of a\nlarge number of blacklisted speakers. It is a form of multi-target speaker\ndetection based on real-world telephone conversations. Data recordings are\ngenerated from call center customer-agent conversations. The task is to measure\nhow accurately one can detect 1) whether a test recording is spoken by a\nblacklisted speaker, and 2) which specific blacklisted speaker was talking.\nThis paper outlines the challenge and provides its baselines, results, and\ndiscussions.\n