HyperAI
HomeNewsTutorialsDatasetsEventsWikiSOTA
Search
About
HyperAI

LibriSpeech ASR Corpus

Date

6 years ago

Size

140.02 GB

Organization

Publish URL

www.openslr.org

License

CC BY 4.0

Tags

Foreign universities
customer service
educate
Natural Language Processing

Categories

Speech Recognition
Dataset DownloadMagnet LinkDownload Help

The LibriSpeech ASR corpus was created by Vassil Panayotov with the assistance of Daniel Povey and includes approximately 1000 hours of 16kHz read English speech, as well as 1000 hours of English pronunciation and corresponding text.

The LibriSpeech ASR corpus was released in 2015 by the Center for Excellence in Human Language Technologies at Johns Hopkins University.

Main contributors: Vassil Panayotov and Daniel Povey

Related paper: LibriSpeech: an ASR corpus based on public domain audio books

LibriSpeech ASR corpus.torrent
Seeding 2Downloading 0Completed 1,168Total Downloads 3,037
  • LibriSpeech ASR corpus/
    • README.md
      1.23 KB
    • README.txt
      2.46 KB
      • data/
        • dev-clean.tar.gz
          322.27 MB
        • dev-other.tar.gz
          622.02 MB
        • intro-disclaimers.tar.gz
          1.26 GB
        • md5sum.txt
          1.26 GB
        • original-books.tar.gz
          1.53 GB
        • original-mp3.tar.gz
          83.45 GB
        • raw-metadata.tar.gz
          83.48 GB
        • test-clean.tar.gz
          83.81 GB
        • test-other.tar.gz
          84.11 GB
        • train-clean-100.tar.gz
          90.06 GB
        • train-clean-360.tar.gz
          111.53 GB
        • train-other-500.tar.gz
          140.02 GB