The 2015 Signal Separation Evaluation Campaign
Abstract
In this paper, we report the 2015 community-based Signal Separation Evaluation Campaign (SiSEC 2015). This SiSEC consists of four speech and music datasets including two new datasets: " Professionally produced music recordings " and " Asynchronous recordings of speech mixtures ". Focusing on them, we overview the campaign specifications such as the tasks, datasets and evaluation criteria. We also summarize the performance of the submitted systems.
Domains
Signal and Image Processing| Origin | Files produced by the author(s) |
|---|
Loading...