---
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
dataset_info:
  features:
  - name: id
    dtype: string
  - name: audio
    dtype: audio
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 5147836328188.375
    num_examples: 10808037
  - name: valid
    num_bytes: 707337368.268
    num_examples: 16393
  download_size: 5148534648325
  dataset_size: 5148543665556.643
---

# Multilingual LibriSpeech (MLS)

This repository is a Hugging Face mirror of the **Multilingual LibriSpeech (MLS)** dataset, originally released on [OpenSLR (SLR94)](https://www.openslr.org/94/).

---

## 📜 License

The dataset is released under the **Creative Commons Attribution 4.0 International (CC BY 4.0)** license.
See the [LICENSE](https://creativecommons.org/licenses/by/4.0/) for details.

---

## 📖 Citation

If you use this dataset, please cite:

```bibtex
@article{Pratap2020MLSAL,
  title = {MLS: A Large-Scale Multilingual Dataset for Speech Research},
  author = {Vineel Pratap and Qiantong Xu and Anuroop Sriram and Gabriel Synnaeve and Ronan Collobert},
  journal = {ArXiv},
  year = {2020},
  volume = {abs/2012.03411}
}
```