Skip to content

Releases: hearbenchmark/hear-preprocess

v2021.1.1

08 Jun 03:05
610d864
Compare
Choose a tag to compare

This version is a minor update containing changes to URLs and language following the renaming of GitHub organization from neuralaudio to hearbenchmark

What's Changed

Full Changelog: v2021.1.0...v2021.1.1

v2021.1.0

21 Jan 18:40
444c645
Compare
Choose a tag to compare

Version used for HEAR 2021 NeurIPS Evaluations

This version:

  • Adds all the secret task config files from HEAR 2021
  • Adds support for TensorFlow datasets
  • Adds split_mode and max_task_duration_by_split in task_config
  • Now supports k-fold splits in addition to train/val/test splits
  • Adds support for rar files
  • Validation for config files

v2021.0.5

13 Sep 18:44
a9722ac
Compare
Choose a tag to compare
  • Open dataset preprocessing, as used in the leaderboard
  • Substantial code cleanup and re-factoring

v2021.0.1

02 Sep 12:39
Compare
Choose a tag to compare
  • Migrated preprocessing code from hear-eval-kit
  • office_sounds is based audio and labels DCASE2016 Task 2, but with a completely different train/val/test split.
  • speech_commands includes the full test set.