Login / Signup

An empirical study of weakly supervised audio tagging embeddings for general audio representations.

Heinrich DinkelZhiyong YanYongqing WangJunbo ZhangYujun Wang
Published in: CoRR (2022)
Keyphrases
  • weakly supervised
  • relation extraction
  • dimensionality reduction
  • superpixels
  • object recognition
  • pairwise
  • topic models
  • object class