spear_dataset

# SPEAR assembly pool — source datasets Namespace: `gaochengxia` | Dataset: `spear_dataset` Pack mode: `tar` (ModelScope file limit: 100000) Uploaded in ascending size order: | # | Archive | Local source | Size | |---|---------|--------------|------| | 1 | `sources/archives/ravdess.tar.gz` | `RAVDESS` | 1.1G | | 2 | `sources/archives/esc50.tar.gz` | `esc50` | 1.4G | | 3 | `sources/archives/fleurs.tar.gz` | `fleurs` | 3.4G | | 4 | `sources/archives/common_voice.tar.gz` | `cv-corpus-26.0-2026-06-12` | 6.2G | | 5 | `sources/archives/nonverbaltts.tar.gz` | `NonverbalTTS` | 6.8G | | 6 | `sources/archives/dailytalk.tar.gz` | `dailytalk` | 6.9G | | 7 | `sources/archives/expresso.tar.gz` | `expresso` | 8.9G | | 8 | `sources/archives/ami.tar.gz` | `ami_ihm` | 16.8G | | 9 | `sources/archives/musan.tar.gz` | `musan` | 22.1G | | 10 | `sources/archives/meld.tar.gz` | `meld` | 30.4G | | 11 | `sources/archives/psc.tar.gz` | `ParaSpeechCaps` | 31.5G | | 12 | `sources/archives/emolia__audio.tar.gz` | `emolia-balanced-5M-subset` | 61.3G | | 13 | `sources/archives/voxpopuli.tar.gz` | `voxpopuli` | 96.3G | See `spear_dataset_code/dataset_labels_mapping.md` for label mappings.