Published March 26, 2021 | Version v1

Features for QbE-STD: wav2vec 2.0 LibriSpeech 960 Transformer Layer 11 (Archive III: top performing features)

  • 1. Stanford University
  • 2. University of Groningen

Description

Best performing features extracted from audio using Transformer layer 11 of the wav2vec 2.0 LibriSpeech 960h on all 10 datasets (eng-mav, gbb-lg, gbb-pd, gos-kdl, gup-wat, mwf-jm, pjt-sw01, wbp-jk, wrl-mb, wrm-pd).

Files

w2v2-T11_eng-mav.zip

Files (3.6 GB)

Name Size
md5:cf875cbf9945193bd32c859c6782283c
963.7 MB Preview Download
md5:f56b473a1cf951c4ebe6dc6c34dc4cf9
584.5 MB Preview Download
md5:dce00eedf4721df6e7fd11c427eeec4b
251.9 MB Preview Download
md5:842dc4651ab34e6bb215203ee9f10050
286.0 MB Preview Download
md5:eff45ed1b0d0ede5b220c538b43e3059
399.3 MB Preview Download
md5:5324cff90c74f4c5c1291d3db5dd5019
172.2 MB Preview Download
md5:314da67402ca4ab03c03d7ef5a502f7b
465.4 MB Preview Download
md5:b7a927d8f424f862f1ddb52ae2f76520
80.4 MB Preview Download
md5:ea8432b91ab57b051e1bf865df49e710
140.6 MB Preview Download
md5:e7ec74288b39bbd5d671f12e3c667fe9
297.4 MB Preview Download