English

Environment Classification via Blind Roomprints Estimation

Audio and Speech Processing 2023-01-27 v3 Sound

Abstract

In this paper we present a novel approach for environment classification for speech recordings, which does not require the selection of decaying reverberation tails. It is based on a multi-band RT60 analysis of blind channel estimates and achieves an accuracy of up to 93.6% on test recordings derived from the ACE corpus.

Keywords

Cite

@article{arxiv.2209.07196,
  title  = {Environment Classification via Blind Roomprints Estimation},
  author = {Malte Baum and Luca Cuccovillo and Artem Yaroshchuk and Patrick Aichroth},
  journal= {arXiv preprint arXiv:2209.07196},
  year   = {2023}
}
R2 v1 2026-06-28T01:21:12.674Z