VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretation

We introduce VoxPopuli, a large-scale multilingual corpus providing 100K hours of unlabelled speech data in 23 languages. It is the largest open data to date for unsupervised representation learning as well as semi-supervised learning. VoxPopuli also contains 1.8K hours of transcribed speeches in 16 languages and their aligned oral interpretations into 5 other languages totaling 5.1K hours. We provide speech recognition baselines and validate the versatility of VoxPopuli unlabelled data in semi-supervised learning under challenging out-of-domain settings. We will release the corpus at https://github.com/facebookresearch/voxpopuli under an open license.

PDF Abstract ACL 2021 PDF ACL 2021 Abstract

Datasets


Introduced in the Paper:

VoxPopuli

Used in the Paper:

LibriSpeech Common Voice Europarl-ST

Results from the Paper


 Ranked #1 on Speech Recognition on Common Voice Spanish (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Speech Recognition Common Voice French VoxPopuli-50K Test WER 9.6% # 1
Speech Recognition Common Voice German VoxPopuli Test WER 7.8% # 3
Speech Recognition Common Voice Spanish VoxPopuli-50K Test WER 10.0% # 1

Methods


No methods listed for this paper. Add relevant methods here