LibriSpeech

URI: https://gptkb.org/entity/LibriSpeech
GPTKB entity
Statements (31)
Predicate Object
gptkbp:instanceOf gptkb:dataset
gptkbp:citation Vassil Panayotov, Guoguo Chen, Daniel Povey, Sanjeev Khudanpur. "Librispeech: an ASR corpus based on public domain audio books." ICASSP 2015.
gptkbp:contains transcriptions
read speech
gptkbp:createdBy gptkb:Vassil_Panayotov
gptkbp:devClean 5.4 hours
gptkbp:devOther 5.3 hours
gptkbp:duration approximately 1000 hours
gptkbp:format FLAC audio
gptkbp:gender balanced male and female
https://www.w3.org/2000/01/rdf-schema#label LibriSpeech
gptkbp:language English
gptkbp:license CC BY 4.0
gptkbp:numberOfSpeakers 2484
gptkbp:releaseYear 2015
gptkbp:samplingRate 16 kHz
gptkbp:separates gptkb:train
examination
dev
gptkbp:source gptkb:LibriVox_audiobooks
gptkbp:testClean 5.4 hours
gptkbp:testOther 5.1 hours
gptkbp:trainClean100 100 hours
gptkbp:trainClean360 360 hours
gptkbp:trainOther500 500 hours
gptkbp:transcriptionFormat plain text
gptkbp:url http://www.openslr.org/12/
gptkbp:usedFor automatic speech recognition
gptkbp:bfsParent gptkb:Speech_Recognition
gptkb:Wav2Vec_2.0
gptkbp:bfsLayer 6