Skip to content

LibriMix

Speech SeperationEnglish

LibriMix is a speech seperation-focused dataset in English that provides 400+ Hours labeled examples.

About LibriMix

Dataset is used for speech source separation in noisy environments. It is derived from LibriSpeech signals (clean subset) and WHAM noise. It offers a free alternative to the WHAM dataset and complements it.

Details

Task
Speech Seperation
Language
English
Format
n/a
Rows / instances
400+ Hours
Creator
Cosentino et al.
Year
2020
Download Paper

Related Speech Seperation datasets

FAQ